Citation
Colorado GEAR UP program design and its effect on student perceptions of postsecondary education

Material Information

Title:
Colorado GEAR UP program design and its effect on student perceptions of postsecondary education
Creator:
Mendelsberg, Scott
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
1 electronic file. : ;

Subjects

Subjects / Keywords:
College preparation programs -- Colorado ( lcsh )
Education, Higher ( lcsh )
College preparation programs ( fast )
Education, Higher ( fast )
Colorado ( fast )
Genre:
non-fiction ( marcgt )

Notes

Review:
Gaining Early Awareness and Readiness for Undergraduate Programs (GEAR UP) is a federal program aimed at equalizing access to higher education for low-income students. GEAR UP, created in 1998, attempts to provide information and support about higher education to students beginning no later than the seventh grade while promoting lasting partnerships among school districts, two and four year colleges, career and technical schools, and other entities to operate GEAR UP projects. This study focuses on the Colorado GEAR UP (CGU) program and sets the stage for future evaluation of programmatic design, implementation, and outcomes. A logic model was created to evaluate the design and implementation of the program, and to establish that program outcomes were as a result of the design and implementation. The logic model is a critical piece to this puzzle. In researching other pre-collegiate programs, it became clear that measurable outcomes were not used and that evaluations consisted mainly of qualitative data and anecdotal accounts. The logic model is used here to help guide the direction of the program with use of specific measurable outcomes and strategies of how to achieve those outcomes. It was a goal of the CGU program evaluation and its effect on student perception of postsecondary education to provide specific strategies to increase college access for low-income students and to show quantitative results. Quantitative data was collected from participants in 23 middle and high schools in Colorado related to college aspirations, college knowledge, and college course participation. The data reveals that students who participated in CGU from 2006-2011 have higher education aspirations, possess a more thorough understanding of college requirements, and participate in a college course earlier in their high school career than their peers.
Thesis:
Thesis (Ph.D.)--University of Colorado Denver. Public affairs
Bibliography:
Includes bibliographic references.
General Note:
School of Public Affairs
Statement of Responsibility:
by Scott Mendelsberg.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
858281109 ( OCLC )
ocn858281109

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

COLORADO GEAR UP PROGRAM DESIGN AND ITS EFFECT ON STUDENT PERCEPTIONS OF POSTSECONDARY EDUCATION By Scott Mendelsberg B.S., University of Northern Colorado, 1989 M.A., University of Phoenix, 1998 A thesis submitted to the Faculty of the Graduate Sch ool of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Public Affairs 2012

PAGE 2

ii This thesis for the Doctor of Philosophy degree by Scott Mendelsberg has been approved for Public Affairs by Dr. Paul Teske, Chair Dr. Kelly Hupfeld, Advisor Dr. Robert Reichardt Dr. Christine Johnson Date: March 7, 2012

PAGE 3

iii Mendelsberg, Scott (Ph.D Public Affairs) Colorado Gear Up Program Design and Its Effect on Student Perceptions of Postsecondary Educati on Thesis directed by Dr. Paul Teske Gaining Early Awareness and Readiness for Undergraduate Programs (GEAR UP) is a federal program aimed at equalizing access to higher education for low income students. GEAR UP, created in 1998, attempts to provide info rmation and support about higher education to students beginning no later than the seventh grade while promoting lasting partnerships among school districts, two and four year colleges, career and technical schools, and other entities to operate GEAR UP pr ojects. This study focuses on the Colorado GEAR UP (CGU) program and sets the stage for future evaluation of programmatic design implementation, and outcomes. A logic model was created to evaluate the design and implementation of the program, and to esta blish that program outcomes were as a result of the design and implementation. The logic model is a critical piece to this puzzle. In researching other pre collegiate programs, it became clear that measurable outcomes were not used and that evaluations con sisted mainly of qualitative data and anecdotal accounts. The logic model is used here to help guide the direction of the program with use of specific measurable outcomes and strategies of how to achieve those outcomes. It was a goal of the CGU program ev aluation and its effect on student perception of postsecondary education to provide specific strategies to increase college access for low income students and to show quantitative results.

PAGE 4

iv Quantitative data was collected from participants in 23 middle and high schools in Colorado related to college aspirations, college knowledge, and college course participation. The data reveals that students who participated in CGU from 2006 2011 have higher education aspirations, possess a more thorough understanding of college requirements, and participate in a college course earlier in their high school career than their peers. The form and content of this abstract are approved. I recommend its publication. Approved: Dr. Paul Teske

PAGE 5

v TABLE OF CONTENTS Chapter I. INTRODUCTION ................................ ................................ ................................ ........... 1 Significance of the Study ................................ ................................ ........................... 3 Statement of the Problem ................................ ................................ ........................... 5 Logic Modeling to Understand Program Design, Program Implementation, and Program Outcomes ................................ ................................ ............................. 6 Research Questions ................................ ................................ ................................ ... 7 Definition of Terms ................................ ................................ ................................ ... 9 Limitations of the Study ................................ ................................ ............................ 11 Organization of t he Study ................................ ................................ ......................... 12 II. LITERATURE REVIEW ................................ ................................ .............................. 14 Introduction ................................ ................................ ................................ ................ 14 The Logic Model and Program Evaluation ................................ ................................ 15 Student Preparation, Articulation, and Completion ................................ ................... 18 Student Barriers to Post Secondary Enrollment ................................ ......................... 20 The Role of Guidance Counselors in Student College Enr ollment ........................... 25 Gaining Early Awareness and Readiness for Undergraduate Programs .................... 2 8 Implementation of GEAR UP in Colorado ................................ .......................... 28 Dual Enrollment Programs ................................ ................................ .................. 29 The Importance of CGU P CAs ................................ ................................ ............ 32 Conclusion ................................ ................................ ................................ ................. 33 III. METHODOLOGY ................................ ................................ ................................ ....... 34 Restatement of the Problem ................................ ................................ ....................... 34

PAGE 6

vi Logic Model ing to Understand Program Design, Program Implementation, and Program Outcomes ................................ ................................ ............................. 34 Research Questions ................................ ................................ ................................ .... 35 Research Design ................................ ................................ ................................ ......... 37 Data Collection ................................ ................................ ................................ .......... 42 Validity ................................ ................................ ................................ ...................... 44 Data Analysis ................................ ................................ ................................ ............. 46 IV. FINDINGS ................................ ................................ ................................ .................... 4 7 Introduction ................................ ................................ ................................ ................ 4 7 CGU Logic Model Design and Implementation ................................ ....................... 48 Lo gic Model Design ................................ ................................ ............................ 50 Resource Inputs ................................ ................................ .............................. 50 Financial Resources ................................ ................................ ................. 50 Organizational Resources ................................ ................................ ........ 50 Community Resources ................................ ................................ ............. 5 1 Private Educational Services ................................ ................................ .... 5 1 Student Activities ................................ ................................ ........................... 5 1 Academic Counseling ................................ ................................ .............. 5 1 College Visits ................................ ................................ ........................... 5 2 PCA Activities ................................ ................................ ............................... 5 3 Training/Retreats ................................ ................................ ...................... 5 3 Staff Bon uses/I ncentives ................................ ................................ .......... 5 4 Annual Conferences ................................ ................................ ................. 5 5 Professional Development ................................ ................................ ....... 5 5

PAGE 7

vii Outputs ................................ ................................ ................................ ........... 5 8 Student Knowledge of College Requirements ................................ ......... 5 8 Student Knowledge of Financial Aid ................................ ....................... 5 8 I mprove Grades and Test Scores ................................ ............................. 5 9 Motivation to Attend College ................................ ................................ .. 5 9 Outcomes ................................ ................................ ................................ ....... 60 Increase Postse condary Aspirations, Awareness, and Knowledge .......... 60 Enhanced Academi c Achievement and Development ............................. 61 Increase d Postsecondary Participation ................................ ..................... 61 Impacts ................................ ................................ ................................ ........... 61 Sh ort Term Benefits to Students ................................ .............................. 61 Long Term Benefits ................................ ................................ ................. 62 Logic Model Implementation ................................ ................................ .............. 64 Research Question I ................................ ................................ ................................ ... 66 CGU vs. Control ................................ ................................ ................................ .. 66 High School Asp i ration and College Aspiration ................................ ................. 6 7 High School Aspiration ................................ ................................ .................. 6 8 High School Aspiration 8th Grade ................................ ........................ 6 9 High School Aspiration 1 1th Grade ................................ ...................... 7 4 CGU Graduation Rates ................................ ................................ .................. 7 9 College Aspiration ................................ ................................ ......................... 81 College Asp iration 8th Grade ................................ ............................... 85 College Aspiration 11th Grade ................................ ............................. 90

PAGE 8

viii CSA P Rates and College Enrollment ................................ ............................ 9 4 CGU vs. N ational Data ................................ ................................ ........................ 9 7 Research Question II ................................ ................................ ................................ .. 9 8 CGU vs. National Data ................................ ................................ ........................ 9 8 PCA Relationship College Ability ................................ ................................ .... 9 9 PCA Relationship College Ability 8th Grade ................................ .......... 102 PCA Relationship College Ability 11th Grade ................................ ........ 10 6 GEAR UP Retentio n Rates and Course Performance ................................ .......... 1 11 Research Question III ................................ ................................ ................................ 11 9 CGU vs. Control ................................ ................................ ................................ .. 11 9 High School Class Requirements ................................ ................................ ......... 11 9 High School Class Requirements 8th Grade ................................ ............... 1 21 High School C lass Requirements 11th Grade ................................ ............. 1 25 Coll ege Entrance Requirements ................................ ................................ ........... 12 9 College Ent rance Requirements 8th Grade ................................ ................. 1 31 College Entr ance Requirements 11th Grade ................................ ............... 1 34 CGU Remedial Course Rates ................................ ................................ ......... 1 39 V. DISCUSSION ................................ ................................ ................................ ................ 141 Introduction ................................ ................................ ................................ ................ 141 Restate ment of the Research Questions ................................ ................................ ..... 143 Research Question I ................................ ................................ ................................ ... 144 Research Question II ................................ ................................ ................................ .. 14 7 Research Question III ................................ ................................ ................................ 149

PAGE 9

ix Study Limitations ................................ ................................ ................................ 151 Implications ................................ ................................ ................................ .......... 152 Suggestions for Further Research ................................ ................................ ........ 156 Learning from the Dissertation Process ................................ ............................... 158 Conclusion ................................ ................................ ................................ ........... 160 REFERENCES ................................ ................................ ................................ ...................... 162 APPENDI X A Survey Questions ................................ ................................ ................................ ..... 173

PAGE 10

x LIST OF TABLES Table I.1 Colorado State High School Graduation Rates by Year and Demographic Group ................................ ................................ .............................. 5 III.1 Survey Questions Labels Used to Answer Res earch Questions I, II, and III .............. 39 III.2 Study Popula tions for CGU Participants and Control Group Members ...................... 44 IV.1 Research Questions and Sample Sizes ................................ ................................ .......... 47 IV 2 Descriptive Statistics Research Question I, High School Aspiration ....................... 6 8 IV. 3 Independent Samples T Test Research Question I, H.S. Aspiration .................... 6 9 IV. 4 Model Summary Research Question I, High School Aspiration 8 th Grade ........... 70 IV. 5 ANOVA Research Question I, High School Aspiration 8 th Grade .......................... 70 IV. 6 Coefficients Research Question I, High School Aspiration 8 th Grade .................. 71 IV. 7 Model Summa ry Research Question I, High School Aspiration 11 th Grade ........ 7 4 IV. 8 ANOVA Research Question I, High School Aspiration 11 th Grade ...................... 7 4 IV. 9 Coefficients Research Question I, High School Aspiration 11 th Grade ................ 7 5 IV. 10 CGU 9 th Grade and 12 th Gr ade Graduation Rates ................................ ...................... 7 9 IV. 1 1 CGU Participating H.S. Graduation Rate and CGU Participant Graduation Rate .... 81 IV. 1 2 Descriptive Statistics Research Question I, College Aspiration ............................ 83 IV. 1 3 Independent Samples Test Research Question I, College Aspir ation .................... 84 IV. 1 4 Model Summary Research Question I, College Aspiration 8 th Grade ................. 85 IV. 1 5 ANOVA Research Question I, College Aspiration 8 th Grade ............................. 85 IV. 1 6 Coefficients Research Question I, College Aspiration 8 th Grade ........................ 86 IV. 1 7 Model Summa ry Research Question I, College Aspiration 11 th Grade ............... 90

PAGE 11

xi IV. 1 8 ANOVA Research Question I, College Aspiration 11 th Grade ........................... 90 IV. 1 9 Coefficients Research Question I, College Aspiration 11 th Grade ...................... 91 IV. 20 CGU Participant Enrollment Type and C O Student Assessment Program Completion ................................ ................................ ................................ ................. 96 IV. 2 1 Descriptive Stat s Research Question II, PCA Relationship College Ability ....... 9 9 IV. 2 2 I ndependent Samples Test Research Question II, PCA Relationship College Ability ................................ ................................ ................................ ........................ 101 I V. 2 3 Model Summary Research Question II, PCA Relationship College Ability 8 th Grade ................................ ................................ ................................ .......................... 102 IV. 2 4 ANOVA Research Question II, PCA Relationship College Ability 8 th Grade ................................ ................................ ................................ ......................... 102 IV. 2 5 Coefficients Research Question II, PCA Relationship College Ability 8 th Grade ................................ ................................ ................................ .......................... 103 IV. 2 6 Model Summary Research Question II, PCA Relationship College Ability 11 th Grade ................................ ................................ ................................ ................... 10 6 IV. 2 7 ANOVA Research Question II, PCA Relationship College Ability 11 th Grade ................................ ................................ ................................ ................... 10 6 IV. 2 8 Coefficients Research Question II, PCA Relationship College Ability 11 th Grade ................................ ................................ ................................ ................... 10 7 IV. 2 9 CGU Participant and Full College Population Retention Rates ................................ 1 12 IV. 30 CGU Participant and Full College Population Course Performance Rates ............... 1 16 IV. 3 1 Descriptive Statistics of Research Question III High School Class Requirements 11 9 IV. 3 2 Independent Samples Test of Research Question III H .S. Class Requirements ...... 1 20

PAGE 12

xii IV. 3 3 Model Summary Research Question III H igh School Class Requirements 8 th Grade ................................ ................................ ................................ ..................... 1 21 IV. 3 4 ANOVA Research Question III High School Class Requirements 8 th Grade ... 1 21 IV. 3 5 Coefficients Research Question III H .S. Class Requirements 8 th Grade .......... 1 22 IV. 3 6 Model Summary Research Question III H.S. Class Requirements 11 th Grade .. ................................ ................................ ................................ ........................ 125 IV. 3 7 ANOVA Research Question III H.S. Class Requirements 11 th Grade .............. 1 25 IV. 3 8 Coefficients Research Question III H.S. Class Requirements 11 th Grade ........ 12 6 IV. 3 9 Descriptive Statistics Research Question III, College Entrance Requirements ..... 12 9 IV. 40 Independent Samples Test Research Question III, College Entrance Requirements ................................ ................................ ................................ ............. 1 30 IV. 4 1 Model Summary Research Question III College Entrance Requ irements 8 th Grade ................................ ................................ ................................ ..................... 1 31 IV. 4 2 ANOVA Research Question III College Entrance Requirements 8 th Grade ...... 1 31 IV. 4 3 Coefficients Research Question III, Survey Question College Entrance Requirements 8 th Grade ................................ ................................ .......................... 1 32 IV. 4 4 Model Summary Research Question III, Survey Question College Entrance Requirements 11 th Grade ................................ ................................ ........................ 1 34 IV. 4 5 ANOVA Research Question III, Survey Question College Entrance Requirements 11 th Grade ................................ ................................ ........................ 1 35 IV. 4 6 Coefficients Research Qu estion III, Survey Question College Entrance Requirements 11 th Grade ................................ ................................ ........................ 1 36 IV.4 7 CGU Participant College Remediation Needs During First Year of College ............ 1 40

PAGE 13

xiii V.1 Research Questions Sample Sizes and Beta ................................ ................................ 141 V. 2 Total Enrollments Based on Ye ars in CGU ................................ ................................ ... 154 V.3 Cost Per Enrollment in CGU ................................ ................................ ........................ 155 V.4 Economic Cost Benefit Analysis of CGU ................................ ................................ ..... 156

PAGE 14

xiv LIST OF FIGURES Figure IV.1 Logic Model Flow Chart ................................ ................................ ............................. 49 IV. 2 CGU versus Control Group: Do You Think That You Will Graduate From High School? ................................ ................................ ................................ ..... 6 7 IV.3 CGU versus Control Group: Do You Think That You Will Go On For Further Education? ................................ ................................ ................................ 83

PAGE 15

1 CHAPTER 1 INTRODUCTION As the United States produces fewer and fewer students prepared to contribute to an ever increasing global economy, too many areas of education hold to past practices restricting postsecondary access to select students who meet narrow academic thresholds. As President Barack Obama said in his speech to Congress on February 24, 2009, it is imperative now, more than ever, that mo re students achieve post secondary education. According to a report by the American Council on Education, using data collected by the Department of Education and the Census Bureau, educational attainment in the United States is strong only with the student demographic with which th e country has always done well, including Anglo and Asian students from h igh socio economic backgrounds (American Council on Education, 2009). It is the rest of our population that is struggling. Although some studies indicate tha t there are some small decreases in the achievement gap in regard to graduation rates and post secondary matriculation rates of underserved students, the progress is minimal and slow. As of 2008, Colorado was ranked 40th nationally in college participation rates for students of low income, at risk families. This longitudinal study was conducted by Postsecondary Educational Opportunity in 2008, a public policy organization that emphasizes education. Understanding the pervasive problems of low performance an d slow academic achievement on the part of minorities and students from low income households remains a problem of great national concern (House, 2008). Unfortunately, education as an entity is deep rooted in standard practices. For a generation now, our e conomy has been creating more opportunities for the productive use

PAGE 16

2 of highly developed cognitive skills then there are people to take advantage of them. Today, while there is a recession effecting broad sections of industry, there are still employment opp ortunities for individuals with postsecondary credentials. However, a growing number of labor experts are raising concerns that workers in the United States lack the critical skills needed to be employable in the current constricted workplace (Watanabe, 20 08). Public schools clearly have to make adjustments to meet this concern. Gaining Early Awareness and Readiness for Undergraduate Programs (GEAR UP) is a federal grant funded by the U.S. Department of Education (DOE). Colorado GEAR UP (CGU) is a specif ic statewide GEAR UP grant managed by the Colorado million dollars each year from DOE, of which 50% is placed in a college scholarship fund. The remaining grant monies fu nded an operating budget that covered a statewide team of site coordinators and on site programs. The Colorado Paradox is the dichotomy in Colorado in which the state has one of the most highly educated work forces in the country fueled by nonresidents mo ving into the state, but its college going rate is below the national average, particularly among p reparing low income Colorado students to meet the high expectations for colle ge admission and graduation, thus helping to level the playing field for the neediest of students and families. The majority of GEAR UP students are from families in which no one has ever attended college. Of the 750,000 students attending Colorado Public Schools, some 250,000 qualify for the National School Lunch Program, better known as Free or Reduced Lunch. CGU provides students and families with the infor mation,

PAGE 17

3 resources and academic support necessary to finish high school and gain admission to col lege. The goal of CGU is to reverse the Colorado Paradox by providing college access and support internally and through public and private partnerships. Significance of the Study The United States continues to slip in education compared to other countri es in the world. This impacts the country in many negative ways. In Colorado it is no different. 20 education system faces a host of serious systemic problems that need to f students necessary for the state to be competitive in the 21 st century. In the coming decades, demographic shift to an older and more ethnically diverse population. Col orado already relies too heavily on importing educated citizens to fill jobs for which native citizens are not qualified. By th e end of this decade, nearly 70% of jobs in Colorado will require at least some higher education. ne is broken; too few students graduate from high school, and of the students that make it to college, too many need remediation and do not end up graduating with a degree. Even for the students that make it to college, the chances of them graduating in f our or six years is much too small considering the time and investment that is made. The following findings highlight these educational shortcomings within the state: 1) Colorado has the 2 nd highest majority minority achievement gap for higher education at higher education is funded at the 48 th lowest in the country on a per student basis; 3) rate is 73. 9% (60.1% for Hispanic students,

PAGE 18

4 66.4% for black students, and 81 .1% f or white students); 4) only 22% Colorado 9 th graders end up graduating college within 6 years; 5) 28.5% system of higher educati on require remediation; 6) 52.7% of community college students require remediation; 7) gradu ation rates at four year researc h institutions range from 31 42% in four years, and 59 73% in six years, depending on the institution; 8) at four year regional institutions, the aver age graduation rate is about 14% in four years and 37% in six years; and 9 ) at community colleges, the aver age graduation rate is about 25% in two years and the same percentage for four years. Table I.1 highlights Colorado high school graduation rate changes from 2009 to 2011 as found in the Colorado Department of Education gra duate file. This file does not account for students leaving the Colorado school system. CGU was required to serve students attending low income schools, and it follows that the largest portion of CGU two of the lowest demographics listed in the table.

PAGE 19

5 Table I.1 Colorado State High School Graduation Rates by Year and Demographic Group Statement of the Problem This quantitative research study discovered how CGU was designed and implemented in an effort to improve student knowledge of collegiate pathways, increase their postsecondary aspirations, and utilize pre collegiate advisors (PCAs) to lead students to pursue postsecondary degrees. CGU is a federally funded statewide program that served low income middle and high school students. This program effectively met and exceeded the few broad federal grant goals established to assure the program was successful, including increasing student academic performance and preparation for Student Demographics 2009 2010 State Totals (All Students) 74.6 72.4 American Indian or Alaska Native 55.9 50.1 Asian 85.7 82.4 Black or African American 64.3 63.7 H ispanic or Latino 57.8 55.5 White 82.3 80.2 Native Hawaiian or Other Pacific Islander n/r n/r Two or More Races n/r n/r Male 71.4 68.7 Female 78 76.3 Students With Disabilities 64.3 52 Limited English Proficient 53.3 49.2 Economically Disadvantaged 61.2 58.9 Migrant 58.3 53.8 Title I 45.8 47.8 Homeless 56.2 48.1 Gifted and Talented 91.6 92.2

PAGE 20

6 postsecondary education, increasing student high school graduation and enrollment in postsecondary education, and increasing student and family knowledge of postsecondary education options, preparation, and financing. In a ddition, CGU implemented numerous enhanced opportunities that assisted students on their path into postsecondary education. Those opportunities were not required by the grant, and were the result of the utilization of a logic model to create and oversee pr ogrammatic services. Knowing CGU sought to provide enhanced options for students, is there a logical connection between the program design, program implementation, and program outcomes? In turn, how did these programmatic services impact students understan ding of the benefits of earning a postsecondary degree and their aspirations to do so in comparison to their peers not in the CGU program? Logic Modeling to Understand Program Design, Program Implementation, and Program Outcomes To assure the impact of pr ogrammatic services provided by CGU were effectively designed and implemented, and that the outcomes of the program can be attributed to these elements, an outcomes approach logic model was used to evaluate the program and its administration. The logic mod el was used to evaluate resource inputs including financial resources, organizational resources, community resources, and private educational services; activities students completed while participants in CGU as well as activities program staff participate d in such as academic counseling, college visits, trainings/retreats, staff bonuses/incentives, annual conferences, and professional development; outputs created by these activities including student knowledge of college requirements, student knowledge of financial aid, improved grades and test scores, and

PAGE 21

7 motivation to attend college; outcomes of the program including students increased postsecondary aspirations, awareness and knowledge, enhanced academic achievement and development, and increased postseco ndary participation ; and the impact of the program on students including short term and long term benefits. The logic model both evaluated the programs design elements and implementation, and also provides prospective on the results achieved through the ac tivities students completed. These The logic model was developed by reviewing CGU program documents including the gra nt application, annual performance reviews, the GCU database, brochures and marketing materials, financial records, training documents, Colorado statewide data, and information distributed to stakeholders. The study then catalogued the resources and activi ties installed by the program to create the intended outputs, outcomes, and impacts. Following this identification process, causal linkages among program components through the use of the linear, columnar outcomes approach logic model was completed. Within the CGU logic model, arrows show which activities were believed to contribute to specific outcomes. These connections serve as logical assertions about the perceived relationships among program operations and desired results. Research Questions The resear ch questions that guided this study were designed to understand student perceptions and aspirations to enter postsecondary education, and the findings resulting from these questions are supported by programmatic data gathered through the logic model. The q uestions examined improved student educational aspirations, and improved

PAGE 22

8 knowledge and understanding of college pathways. Finally, the questions looked to help understand student perceptions on how CGU influenced their ability to attend postsecondary educa tion and understanding of financial aid. Research Question I: Is there a correlation between student participation in CGU and their educational aspirations? Hypothesis I .1 : Low income students that participated in CGU will demonstrate higher aspirations to graduate from high school over time. Hypothesis I .2 : Low income students that participated in CGU will demonstrate higher aspirations to earn a college degree over time. Hypothesis I .1 and 2 captures the idea that if students are exposed to information about college in a systemic way, their individual aspirations for post secondary education will be enhanced and that will show within survey results. In each of these questions the independent variable is exposure to CGU curriculum and PCAs by the program participants versus compared to the control group not exposed to CGU curriculum and PCAs, and the dependent variables are their responses to survey questions. The rival hypothesis for Research Question I is that CGU does not improve low a spirations to earn a high school or college degree. Research Question II: Is there a correlation between CGU PCA mentoring and student aspirations to pursue postsecondary education? Hypothesis II .1 : Low income students that participated in CGU will be mor e likely to pursue a postsecondary degree due to the strong relationship they developed with their PCA over time.

PAGE 23

9 Hypothesis II 1 addresses the idea that if students have a dedicated professional working with them on a regular basis, the likelihood of the m pursuing a postsecondary relationships with PCAs do not influence their pursuit of a postsecondary degree. Research Question III: Is there a correlation between student participation in CGU and their knowledge and understanding of postsecondary pathways? Hypothesis III .1 : Students aspirations to attend postsecondary education will increase over time. Hypothesis III .1 considers that if students are exposed to inf ormation about college in a way that is systematic with concepts introduced at specific times and revisited regularly, their aspirations to attend postsecondary education will increase as exhibited in their responses to survey questions. In each of these q uestions the independent variable is the implementation of the curriculum and the dependent variables are related to the specific questions. The rival hypothesis for Research Question III is that students do not gain hope in their ability to attend post sec ondary education over time. Definition of Terms Advanced placement includes courses students complete in high school for high school credit culminating in a test to judge their competency in the subject. Based on this score, they are given college credit f or one or more courses. Cohort model is a grouping of GEAR UP students moving through all services and phases of the program together College Level Examination Program (CLEP) tests high school students on content taught during the first two years of colle ge. The tests are usually on content from college

PAGE 24

10 introductory courses, and high school students earning satisfactory scores on these exams receive college credit. Colorado GEAR UP (CGU) is the specific statewide federal GEAR UP grant researched within thi s study. Concurrent enrollment, also known as dual enrollment, allows high school students to earn simultaneous college and high school credit through courses taught at a postsecondary institution (Greenberg, 1989). Courses are typically taught on a commun ity college campus. Dual enrollment also known as concurrent enrollment, offers simultaneous college and high school credit to high school students. Dual enrollment is unique because courses are taught by high school teachers certified to teach both high school and college courses, and occur at the high school within the typical high school day (Education Commission for the States, Center for Community College Policy, 2001). Economically disadvantaged students qualify for either the free or reduced lunch program. Gaining Early Awareness and Readiness for Undergraduate Programs (GEAR UP) is a federally funded advising program that assists low income students on their paths to and through postsecondary education. High school counselor is an individual sala ried by a school district to assist students through their middle and high school careers. Instructors teach courses at the postsecondary level, are not working toward tenure, and may or may not perform research. Additionally, this term will be used to des cribe anyone who teaches at the postsecondary level.

PAGE 25

11 International Baccalaureate Diploma Programme is the high school component of the International Baccalaureate. Students complete six courses in six subject areas for high school credit. Many universities consider these courses equivalent to college courses and provide related credit upon admission. The National School Lunch Program, or Free and Reduced Lunch Program provides children from famili es with incomes at or below 185% of the poverty level with me als while in school. This program also dictates student eligibility for GEAR UP. Persistence generally refers to when a student reenrolls at an institution year after year and reaches graduation. The most commonly researched persistence point is when fres hmen return to an institution for their sophomore year. Persistence can also be defined as the rate at which a class of students continues at the same school the following year. Pre collegiate advisor (PCA) is an individual salaried by CGU to provide pre collegiate services to program participants within a middle or high school. Professors teach at the postsecondary level, are advancing toward tenure, and are responsible for performing research within their field. Teachers are educators at the secondary level. Title I students are identified by the school as failing, or most at risk of failing, to multiple, educationally related, objective criteria established by the scho ol. Limitations of the Study This study focused on the perceptions and aspirations of high school students from local public school districts who were enrolled in CGU. Since each educational level and each environment within those levels have distinct cha racteristics, it is difficult

PAGE 26

12 to extrapolate the findings of this study to other school districts. However, it should be noted that study participants and the schools they attend are representative of low income students and schools around the country. Th is project utilized the priority student approach in which every student in the participating grade levels and schools who were eligible for the free and reduced lunch program and demonstrated poor academic performance was targeted as a project participan t. The schools selected for project participation were chosen based on high numbers of low income students and high numbers of students from groups typically underrepresented in college. This study did not address if the services provided by CGU impacted c ollegiate enrollment rates or perceptions of college attendance by middle or high income student. The students that self selected to be in the program were certainly students with a desire to increase their opportunities through postsecondary education. Based on a conversation with the Director of the Colorado GEAR UP program at that time, Gully Stanford, most students were chosen to participate in the program based on the criteria noted above. Some schools identified specific students to enroll in CGU du e to their performance on C SAP in at least one subject of While most participants were randomly selected on a first come, first served basis at CGU schools, these students were enrolled in the program because o f their academic difficulties in at least one subject, potentially influencing the results and findings of this study. Organization of the Study Chapter 1 provides an overview of the current educational landscape in the United

PAGE 27

13 States and specifically in t he state of Colorado, describing why this study is important. Chapter 2 reviews the research literature on student preparation for college, barriers to attendance, pre collegiate programs, and the use of logic modeling to create and understand organization s. A complete overview of the research methodology follows in Chapter 3 including research instruments, data collection, recording procedures, and analysis of participant answers to interviews. Chapter 4 presents the data collected and an analysis of find ings. Chapter 5 discusses important findings related to the three questions posed by the study, offers implications, suggests recommendations for further research, and makes conclusions.

PAGE 28

14 CHAPTER 2 REVIEW OF THE LITERATURE Introduction The overarching go al of GU was to increase college preparation for low income students. The grant funds programs to work with students through a cohort program over six years beginning in the 7th grade. Early interventions and a scholarship component are two of the requirem ents that must be incorporated into any GEAR UP program. Cabrera, Deil Amen, Prabhu, Terenzini, Lee, and Franklin (2006) consider GEAR UP a comprehensive intervention program. This differentiates GU from previous large scale programs funded at the federal or regional level. Starting in middle school, CGU provided services (counseling, curriculum, dual enrollment opportunities and college visits) to assure that low income students were prepared to attend college after high school graduation. Implementing a p roject the scope of CGU can be challenging. It is important that that there are measureable outcomes and rationale behind results. A logic model is a systematic visual way to present and share the relationship between resources educators have to operate pr ograms, activities educators plan, and changes or results educators hope to achieve (W.K. Kellogg Foundation, 2004). They are a picture of why and how programs are believed to work. In logic model methodology, one is required to describe, share, discuss, a nd improve program theory in words and pictures while developing, implementing, and evaluating a program (W.K. Kellogg Foundation, 2004.) The following literature review was designed to show how CGU addressed long term concerns about the achievement gap an d the postsecondary matriculation gap

PAGE 29

15 between students from varying ethnic and socioeconomic backgrounds. The review covers some of the literature that describes the barriers that minority students face in their pathway to college. The review focused on so me of the strategies that CGU used to close the achievement gap a nd the development of a logic model to capture these strategies and expected outcomes. Finally, this literature review demonstrates how a logic model can help organizations evaluate program s ervices, implement activities and measure success. The Logic Model and Program Evaluation Implementing a project with the scope of CGU can be challenging. It is important that that there are measureable outcomes and rationale behind results. A logic model is a systematic visual way to present and share the relationship between resources educators have to operate programs, activities educators plan, and changes or results educators hope to achieve (W.K. Kellogg Foundation, 2004). They are a picture of why a nd how programs are believed to work. In logic model methodology, one is required to describe, share, discuss, and improve program theory in words and pictures while developing, implementing, and evaluating a program (W.K. Kellogg Foundation, 2004.) Review ing program design using a logic model involves collecting information about program components and outcomes in order to improve program effectiveness, or worth (Patton, 1 997). It is an essential part of program planning and the implementation process, because it provides information about necessary improvements and accountability (Lee & Sampson, 1990). The logic model can be used to reflect the theory of how a project is going to promote change within a specific target system (C.S. Mott, 2005). This study will consider how CGU increased college knowledge, college

PAGE 30

16 aspirations, and college access for low income students in Colorado. Program evaluations using logic models are frequently undertaken to provide pragmatic information to program administrators and managers who are considering changes to elements of their organization (Royse, Thyler, Padgett, & Logan, 2001). In the case of this particular study, the national Departm ent of Education could use the findings to define the model for reviewing of all other state GEAR UP programs. For this study, a n outcomes approach logic model was used to better understand CGU. Outcomes approach logic models are often found in grant prop osals because they are used to explain how the elements of the project design fit together to produce desired outcomes. Outcomes approach logic models often emphasize the following elements: (a) problems, concerns, issues targeted for solution; (b) unique assets the project or organization staff bring to the project design to address the problem; (c) assumptions made from underpinning theory the project strategies are based upon; (d) environmental, social, or contextual conditions that could influence on th e expected outcomes through project activities; (e) project activities based on identified, proven interventions addressing similar problems or issues; and (f) outcomes, results, or changes expected because of the project activities (C.S. Mott, 2005). The logic model of a program such as CGU is a picture of how the organization does its work, including the theory and assumptions of the program. Often tables and flow charts are used to illustrate program dimensions. This can also be a useful training tool f or new employees to understand the underlying principles of the work and the mission of the organization. Although there are several types of logic models, the program logic model will be utilized because it links outcomes (both short and long

PAGE 31

17 term) with program activities and accounts for the theoretical assumptions and principles of the program (W.K. Kellogg Foundation, 2004). It is important that the study of federal educational programs such as GEAR UP have outcomes that are clearly related to the meas urable federal objectives. Using the logic model approach, it is possible to achieve effective programming and offer greater learning opportunities, better documentation of outcomes, and shared knowledge about what works and why (McElvain, 2004). The logic model used in this study evaluated the design of the program, implementation, and linked these elements to outcomes of the program. the Harvard Family Research Project (1999) and t he W. K. Kellogg Foundation (2004). This is a sequential model of actions leading from resources to impact and are usually depicted in 5 steps: (1 ) inputs: resources available to a progra m for use in task completion; (2 ) program activities: processes, tool s, events, technology, and actions that are integral parts o f the program implementation; (3 ) outputs: direct products of program activities that include types, levels, and service targets to be delivere d by the afterschool program; (4 ) outcomes: specific status, and level of functioning (short term outcomes are those achievable within 1 3 years; long term outcomes are those achievable withi n a 4 6 year time frame); and (5 ) impact: fundamentally intended or unintended changes in organizations, communities, or systems as a result of program activities.

PAGE 32

18 Student Preparation, Articulation, and Completion earning potential (NCES, 200 2), and a four year degree creates even greater individual and societal benefits (Grubb, 1999). This message is being heard; almost two thirds of high school graduates enter postsecondary education including community colleges and vocational schools (NCES, 2002). However, Greene and Forster (2003) stated that while students are entering postsecondary education at a high rat e, only 32% of those students are eligible to attend a university with more selective admissions standards. Similar studies, such as one completed by the National Center for Public Policy and Higher Education (NCPPHE), further analyzed college completion statistics. The NCPPHE (2 006) report stated that only 26% of college students are enroll ed after two years, and only 18% graduate with ei contribute to these diminishing results, and preparation for the college experience is seen by educators as the area with the greatest room for improvement. Clearly, preparation and retention improveme nts must be made by secondary and postsecondary institutions alike. Adelman (2006) pointed to a need for greater academic degree is the quality and intensity of their hi gh school curriculum. Students are hearing the message that they must earn a postsecondary degree to become competitive in our knowledge based workforce, and that preparation for earning that degree requires challenging themselves during high school. Why a re national persistence and graduation statistics so low? Lack of preparation for college have been articulated as one possibility: poor instruction quality; lack of

PAGE 33

19 expectations by schools for underrepresented students; and few coursework opportunities to improve critical thinking, writing, and reading skills. Other factors cited by Kirst and Venezia (2001) include grade inflation, a lack of early and high quality college counseling, and subjective student assessment. College preparation for all students i s an immense goal and researchers state that responsibility cannot fall solely on secondary education. Bailey, Hughes, and Karp (2003) argued the open door admission policies of community colleges lead students to believe entrance into the system is simple and progression toward a degree is equally simple. According to Rosenbaum (1998), students understand they can continue schooling with little track record of academic success, but do not realize they will begin in remedial courses, essentially repeating w hat they did not acceptably finish in high school. Some educators argue communication between secondary and postsecondary systems must improve (Orr, 1998; 1999). Without consistent communication among leaders about admission policies, preparation expectat ions, and methods to improve academic success, the gap widens. Kirst and Venezia (2001) asserted that few high school faculty members understand college admission policies and, as a result, cannot articulate the policies to students. According to Zusman (1 999), high school leaders develop negative attitudes toward colleges when they perceive college administrators and professors are dictating curriculum. Conversely, college leaders perceive a lack of effort or poor planning on the part of high schools when they enroll and assist underachieving first year college students. In order to force a stronger relationship between high schools and colleges, Bailey, Hughes, and Karp (2002) suggested implementing a P 16 system with the first two years of

PAGE 34

20 college provid ed in high schools. Since educational boards, funding, budgets, districting, course content, accreditation, and teacher certification all lay on the existing structures supporting secondary and postsecondary systems, reconstructing these systems would prov e extremely difficult. Researchers also emphasize that college preparation expectations must be consistently communicated to K 12 school leaders to tie high school and college policies together (Kirst & Venezia, 2001). Student Barriers to Post Secondary E nrollment Cabrera and La Nasa (2000) noted that students from low income socioeconomic backgrounds are far less likely to have access to information about college enrollment, and this disparity has existed for over three decades. In addition to college adm issions standards financial aid, scholarships, and loan debt impact their likelihood to attend college. Those exposed to financial information about college are more likely to attend (Cabrera & La Nasa, 2000; Alexitch, Kobusses & Stookey, 2004). Lozano, Watt, and Huerta (2009) recognized that low aspirations can be a barrier to college. Increased access to collegiate information allows students to weigh their options after hig h school, and provides them time to mentally prepare for the collegiate transition, which is often stressful. Through focused workshops and one on one sessions, school graduati on to college graduation. For their study of an intervention with the goal of increasing Hispanic student college admission, Marsico and Getch (2009) found barriers included the application process, lack of parental knowledge about high school

PAGE 35

21 requirements lower school administration expectations for Hispanic students, and a lack of support from counselors. While information attainment is vital for students to make informed decisions about their educational choices, relationships with those most likely to provide this information and support their aspirations are also impactful. Cooper and Liou (2007) found that students who did not frequently visit with their counselors had a weaker understanding of rigorous coursework requirements and the benefits of goin g to college. GEAR UP is designed to counter this by building a strong relationship between the PCA and students starting in the 7 th grade. Downs, Martin, Fossum, Martinez, Solorio, and Martinez (2008) recognized that a lack of parental involvement and sup port is a barrier to college enrollment. They developed a six week college knowledge program which tested students and parents before the program to learn their past immersion in college information. The pretest found that a very high percentage of Latino ( a) students and parents had never received information about financial aid, college entrance requirements, careers, or college testing. There are numerous pre collegiate programs throughout the country that serve students from at risk backgrounds that ar e not highly represented at higher education institutions. There is a gap in high school academic achievement, persistence, college attendance, and college degree attainment when viewed in comparison to income levels, the first generation to attend college from their family. Specifically, white students achieve a higher rate of degree attainment than do their African American and Latino counterparts. The review of the literature suggests that there are distinctive chara cteristics as to why students from certain ethnicities do not

PAGE 36

22 attend college. Two interrelated characteristics that highlight difficulties some student subsets have in earning degree are household income and ethnicity. Based on a 2006 report, the median ho usehold inc ome for African Americans is 60% of that of Whites. For Latinos their median income is only 86% of Whites (U.S. Department of Education, 2006). The implication is that students from low income backgrounds lack familial background and the financ ial support to fund entrance into college. Students from ethnic minorities that do gain access to postsecondary institutions are more likely to enroll in certificate programs or associate degree programs rather than degree programs (Ishanti, 200 5; McCarron & Inkelas, 2006; Nunez & Cuccaro Alamin, 1998). Students are also more likely to progress toward a college degree if they enter college directly after graduating high school, is often more difficult for students from low income backgrounds bec ause of financial concerns. Often, there are family and community pressures on students to begin contributing income once they leave high school, forcing some to put off school or not attend. However, those who wait to begin postsecondary work are putting themselves at a disadvantage; Students who begin college right away are more likely to graduate from college within five years compared to students who enter with a break between high school and college (National Science Foun dation, 2002). In 2006, only 5 5% of African Americans and 58% of Latinos enrolled in college during the fall semester af ter spring graduation, while 69% of Whites progressed directly into college (U.S. Department of Education, 2008). While entering college directly is difficult for ma ny students based on familial expectations, those who do begin right away still face formidable challenges. Remedial courses in college present a difficult barrier to overcome, especially for students from

PAGE 37

23 low income backgrounds a nd certain ethnicities. O nly 30% of students who attend public high schools in the United States are college ready by the twelfth grade (The National Center For Public Policy and Higher Education, 2008). While this rate is low, it is even lower for minor ity students. Only 23% of A frican Americans and 20 % of Latinos, compared to 40 % of Whites, are college ready upon high school graduation grade (The National Center for Public Policy and Higher Education, 2008). First, these students began at a disadvantage, having to complete more c ourses than their counterparts who are able to start coursework at the introductory level. Second, because of this added layer of courses, the path toward a degree becomes more expensive. Third, as these students have not reached the level of beginning col lege at the introductory level, it can be argued they have not honed the skills necessary to succeed in college to the degree that their classmates have. Many from low income backgrounds are the first from their family to attend college, labeled as first generation students. Their parents and siblings have not attended college and have little understanding of the skills and knowledge necessary to enroll in and succeed at a postsecondary institution. When looking specifically at aspiring minority college s tudents, conversations with their parents were less about going to college and more about getting a job after high school (Fallon, 1997). This finding highlights one reason it is difficult for first generation students to break the pattern of not progressi ng past a high school education within their family, but also why many low income students choose to work after high school rather than continuing into college. Another study found that one third of aspiring first generation college students who gained ac cess to college never discussed entrance exam preparation with their parents and only 42 %

PAGE 38

24 discussed applying to college (Fallon, 1997). Only one fifth of students whose parents earned a college degree neglected to discuss college e ntrance exam preparation and 61% often discussed applying to college with their parents (Fallon, 1997). Even if low income minority students had taken the same courses in high school, they would be less likely to take test prep courses that could improve their scores (JBHE, 1999). Low income students also reported they cannot depend on their parents to assist them with the financial aid process, which is essential for these students to gain access to postsecondary institutions (De La Rosa & Tierney, 2006; Perna, 2006). Students fro m low income backgrounds are also forced to attend high schools in low income areas, which are often provided fewer resources to manage a larger range of academic and behavioral issues. Economic status affects K 12 school choice, based on neighborhood loca tion. For low income minority students, economic status results in limited access to quality schools (Fuller, 2002). One of the benefits of GEAR UP is additional resources to schools that need them. Case studies and ethnographies have found that schools w ith a low socio economic school designation do not hire and retain the most qualified teachers, or those who test highest on content certification exams (Kahlenberg, 2006; Kozol, 2001). Strong teachers are important in creating student interest in a subjec t, and in instilling the relevance of education in general. Studies have shown that low income minority students are more disengaged than non minority students, especially if they do not view education as relevant to their future (Constantine, Kindaichi, & Miville, 2007). Another barrier for potential first generation college students is the overall lack of rigorous academic preparation. Students from low income backgrounds who aspire to

PAGE 39

25 attain a four year degree had lower scores on cognitive tests, lower g rades, and earned less Carnegie units in math and English compared to students from less challenged economic backgrounds. One cause may be that ethnic minorities are not pushed as consistently to succeed in their courses as their white classmates. It has b een found that white students receive more academic encouragement than non white students. These reasons highlight why many low income students do not enter college, and those that do begin with remedial coursework. The Role of Guidance Counselors in Stu dent College Enrollment According to the American School Counselor Association the roles of guidance counselors include providing services to students, parents, and school staff. The specific duties include curriculum guidance, student educational planning prevention and intervention, group counseling, referrals to support services, promotion of systemic developmental level. The importance of relationships between students and schoo l leaders, including counselors, extends beyond the benefits students receive by learning about college and academic opportunities. School leaders were found to be one of the top three community groups that a foster positive self concept among adolescents (Gibson & Jefferson, 2006). It was also discovered that the context of the school, whether students, employees, and community members view the school in a positive or negative light, informs how school counselors lead (Janson, 2009). Within the framework of CGU, a positive working relationship with school administration is critical, but this relationship must be cultivated so that school leaders view the program as an asset to the school. Although additional resources are always helpful, they do bring wit h them challenges.

PAGE 40

26 Challenges come in the form of office space, computer/phone allocation, adoption of remedial coursework and dual enrollment opportunities. Having school administrative support can make a large difference in implementation. Implementation is largely the charge of the CGU assigned PCAs. Their roles mirror many of the job responsibilities that used to fall under high school counselors. However, the role of the counselor in the high school suffers from ambiguity as more and more responsibilit was to provide college information and career advice for those students that were not prepared or did not show interest in college. Krei and Rosenbaum (2001) found that there are n o prescribed or required counselor activities to advise students that are not planning to go to college. Additional duties of the school counselor include leadership, advocacy, and systemic change efforts (Camizzi, Clark, Yacco, & Goodman, 2009). Counselor s behave in their leadership role in numerous ways (Janson, 2009). As the job becomes more complex and includes more functions, the counselor is potentially seen as providing inadequate information about academic planning and college preparation (Beesley, 2004). In too many cases, the ratio of student to counselor makes it impossible to disseminate all of the information about college access in any meaningful way. In examining stages of college choice, Cabrera and La Nasa (2000) found that low income studen ts with a strong and consistent relationship with their counselor were more likely to attend college. CGU maintains PCA to student ratios at 150 1. The importance of this relationship increases as the number of at risk students increases (Beesley, 2004). W hile the relationship is vital, it is most important that the counselor provides useful and easily understood information through targeted interventions. In a study by Camizzi et al.

PAGE 41

27 (2009), low income students who attended monthly workshops completed cour sework with a higher level of academic rigor than fellow classmates who did not attend, and were more likely to complete FAFSA and scholarship documentation. Alexitch, Kobusses, and Stookey (2004), in study of Canadian high sch ool students found that only 33% of the students reported meeting with a guidance counselor. The study also found that students who met with guidance counselors more often were more educated about admission requirements and funding issues; however, the guidance counselors were provid ing similar information to all students no matter their goals or background. PCAs for CGU are required to meet with every student at least twice per month, which assures students receive college going information and that the information provided is speci fic to their goals. In a study of Mexican American Gulf Coast GEAR UP students by Castillo, Conoley, Cepeda, Ivy, and Archuleta (2010) found family influence, peer influence, school personnel, and student responsibility to be four factors that create a pro college culture. All important elements to creating a school environment where students are expected to go to college. However, the students surveyed expected information to originate through the teachers, not counselors. They also placed responsibility f or the understanding of the benefits of college and the completion of college going material on the high school staff, not themselves. In addition to this being an inappropriate understanding of responsibility by the student, counselors must empower studen ts to feel it is their responsibility to understand their collegiate options.

PAGE 42

28 Gaining Early Awareness and Readiness for Undergraduate Programs GEAR UP is funded through the United State Department of Education (DOE). The overarching goal of GEAR UP is t o increase college preparation for low income students. The grant funds programs to work with students through a cohort program over six years beginning in the seventh grade. Early interventions and a scholarship component are two of the requirements that must be incorporated into any GEAR UP program. Cabrera, Deil Amen, Prabhu, Terenzini, Lee, and Franklin (2006) consider GEAR UP a comprehensive intervention program. This differentiates GEAR UP from previous large scale programs funded at the federal or re gional level. Starting in middle school, it provides all services except academic tutoring to assure that low income students are prepared to attend college after high school graduation. GEAR UP exists and succeeds in impacting student goals regarding coll ege because of its PCAs (PCAs). These individuals serve as another school counselor for participants, and assure they are motivated to progress in school and educated about the path they will take into college. Implementation of GEAR UP in Colorado Nation wide, GEAR UP is generally implemented through a partnership between an institution of higher education and a school district or an individual school. Most GEAR UP grants are administered at single higher education institutions and may have partnerships wi th multiple schools or school districts. The implementation of GEAR UP in the state of Colorado differs from this prevailing model. Colorado administers a single GEAR UP program at the Department of Higher Education. The administration of this grant from a central agency rather than a single higher education institution allows for a broader reach across the state. CGU has identified multiple resources to serve students. A

PAGE 43

29 single institution may not have access to all the resources available without centrali zed administration. As an example, CGU includes participants from 23 schools from geographically diverse areas of the state that includes urban, rural and suburban schools. The participants receive information and have access to multiple higher education r esources since there is not a single IHE administering the grant. Collaborations among multiple stakeholders are also at the core of CGU, which were found to be at the core of successful intervention programs for low income students (Camizzi, Clark, Yacco & Goodman, 2009). CGU meets federal program requirements by providing numerous resources to students and the schools they attend. Program services include : 1) providing PCAs to each participating high school 2) a systemic offering of college going curric ulum, 3) college visits, 4) grade monitoring, 5) program/school administrator communication, 6) dual credit courses, 7) a renewable college scholarship, 8) family involvement opportunities, and 9) collaboration with multiple colleges. Although CGU provides comprehensive services, it has a focus on two critical activities designed to increase the college awareness of the participating students, including a concentration on student/counselor relationships, and student enrollment in dual credit programs. Dual enrollment programs. Embracing dual enrollment strategies achieves multiple goals including exposing the students to the college experience and building confidence by demonstrating that college success is achievable. Many first generation students, those w ho would be the first in their family to graduate from college, are intimidated by the entire process. Easing them into it through dual enrollment allows them to take on this challenge while feeling

PAGE 44

30 supported by their high school environment. While these c ourse options are called dual enrollment programs, dual credit programs, and concurrent enrollment programs interchangeably through the literature and by districts and colleges that implement them, they will be referred to as dual enrollment programs for t he purposes of this study. Greenberg (1989) offered a comprehensive list of the benefits to high school students who participate in dual enrollment programs. These results are supported by ce to earn college credit, while potentially decreasing the time to completion of a college degree (Greenberg, quickly than their classmates who did not enroll in credit based transition courses. This particularly affects students from low income backgrounds who already struggle to cover the cost of tuition, fees, books, and housing. Programs like AP and IB have offered courses to high achieving students since the 1950s, b ut it has only been recently that moderately achieving students enrolled in college credit programs. Osborn (1928) and Blanchard (1971) cited curriculum overlap as a second benefit. Rather than complete senior year courses and college introductory courses that overlap, students earn credit at both institutions, saving both time and the frustration of repeating coursework. Courses can be offered at the high school or on the college campus. Another benefit, according to Greenberg (1989) and confirmed by Garto n (2003) and Orr (2002), is savings. Tuition savings assists middle income students who receive smaller financial assistance packages, and alleviates some cost concerns for lower income students. The responsibility for covering annual college expenses, and repaying student loans after graduation, becomes less daunting to students when they are provided an

PAGE 45

31 option to diminish these financial responsibilities upfront. Additionally, students gain confidence in their ability to understand college level coursewor k and succeed in course assessments (Pennington, 2004). Direct experience with the rigor of college coursework assures students they have the intellectual background and the critical thinking skills to succeed in college level courses (Greenberg, 1989). Si nce students learn early what is required to succeed in college, they avoid expensive false starts. (Bailey et al., 2002). of institutions, should not be considered the same as developmental maturity. In many cases, the unsuccessful student is simply unaware of the services and support that they have access to. College readiness can be defined as the level of academic preparation needed for students to earn good grad es in college level courses to progress through sequential requirements (Conley, 2007). Finally, credit school curriculum, (Greenberg, 1989). It is difficult to motivate students to take ri gorous coursework, much less do well in those courses, if they do not believe that they are going to college. Students accessing college courses in high school makes attending college more of a reality to students. Still, dual enrollment programs are relat ively new developments within the education community. Within the last 10 years, there has been a dramatic increase of state level legislation further codifying dual enrollment as a strategy to increase college access to low income and underrepresented pop ulations. Currently, forty two states have either a dual enrollment statute or policy from the state board of education promoting the wide dual enrollment

PAGE 46

32 program began at the University of Color ado Denver. It was called the CU Succeed program. Started in 1990, it has expanded at a generous rate ever since. Specifically in the State of Colorado, the state legislature passed House Bill 09 1319 and Senate Bill 09 285 in May of 2009. It is generally referred to as the Current Enrollment Acts. The purpose of the legislation was to increase dual enrollment programs throughout the state and to improve on the coordination between K 12 and post secondary institutions and to try and ensure financial obligat ions for each while developing a greater accountability. In addition the legislation allowed students to take a fifth year of high school (known as the ASCENT year) to enroll in all college level school districts. Students are responsible for purchasing all books and supplies for the courses for which they are enrolled. The importance of CGU PCAs. Including a CGU specific counselor, or a PCA, in participating middle and high schools creates an opp ortunity for students to frequently visit with an advocate whose primary function is to assure they understand the benefits of a postsecondary education Cabrera et al. (2006) identifies the GEAR UP cohort model as a unique way to work with students, rathe r than working with individuals. Bruce, Getch, and Ziomek Daigle (2009) found that a group counseling approach can increase African American student achievement. The network framework of GEAR UP services increases the social and cultural capital of partici pants, which can reduce some barriers of college attainment (Cabrera et al., 2006; Hewett & Rodger, 2003).

PAGE 47

33 Counselors do not have time to focus on the individual students as much as is needed (Krei & Rosenbaum, 2001; Hewett & Rodgers, 2003). CGU has foun d success by not only limiting student to counselor ratios to 150:1, but by having a proscribed curriculum that is delivered monthly to all students. These lessons cover subjects such as how to compute high school grade point averages, proper preparation f or standardized tests, how to understand transcripts, and what documents must be completed to receive th grade) and expounded upon through high school. Conclusion There are many theories as to why students of color do not attend postsecondary institutions at the same rate as their white peers. There have been a countless amount of pre collegiate programs that were designed to address this very problem. Unfortunate ly, even with all of the attention to this problem, there has been little progress in closing the achievement gap. This literature review attempted to focus on some of the work that has been done to identify the barriers. The review then addressed strategi es that CGU has used to address the problem. Finally, the review focused on how the logic model can be used to help organizations such as CGU, strategize, measure success, and how to use those measures to drive future decisions.

PAGE 48

34 CHAPTER 3 METHODOLOGY Resta tement of the Problem This quantitative research study seeks to discover how CGU used a multi faceted program to implement, administer, and evaluate programmatic opportunities, and to learn te pathways, their postsecondary aspirations, and if their relationships with pre collegiate advisors (PCAs) lead them to pursue postsecondary degrees. CGU is a federally funded statewide program that served low income middle and high school students. This program effectively met and exceeded the few broad federal grant goals established to assure the program was successful. In addition, CGU implemented numerous enhanced opportunities that assisted students on their path into postsecondary education. Those opportunities were not required by the grant, and were the result of the utilization of a logic model to create and oversee programmatic services. Knowing CGU sought to provide enhanced options for students, is there a logical connection between the progr am design, program implementation, and program outcomes?? In turn, how did these programmatic services impact students understanding of the benefits of earning a postsecondary degree and their aspirations to do so in comparison to their peers not in the CG U program? Logic Modeling to Understand Program Design, Program Implementation, and Program Outcomes Using CGU program documents including the grant application, annual performance reviews, the GCU database, brochures and marketing materials, financial r ecords, training documents, Colorado statewide data, and information distributed to

PAGE 49

35 stakeholders, this study evaluated how the program was designed and implemented, and if the results of the program and students perceptions about postsecondary education ca n be attributed to the efficacy of that design and implementation. The logic model was used to review resource inputs including financial resources, organizational resources, community resources, and private educational services; activities students compl eted while participants in CGU as well as activities program staff participated in such as academic counseling, college visits, trainings/retreats, staff bonuses/incentives, annual, conferences, and professional development; outputs created by these activ ities including student knowledge of college requirements, student knowledge of financial aid, improved grades and test scores, and motivation to attend college; outcomes of the program including students increased postsecondary aspirations, awareness and knowledge, enhanced academic achievement and development, and increased postsecondary participation; and the impact of the program on students including short term and long term benefits. The logic model both evaluates the programs design elements and impl ementation, and also provides prospective on the results achieved through the activities students completed. These programmatic results serve to support the findings of Re search Questions The research questions that guided this study were designed to understand student perceptions and aspirations to enter postsecondary education, and the findings resulting from these questions are supported by programmatic data gathered thr ough the logic model. The questions examined improved student educational aspirations, and improved knowledge and understanding of college pathways. Finally, the questions looked to help

PAGE 50

36 understand student perceptions on how CGU influenced their ability to attend postsecondary education and understanding of financial aid. Research Question I: Is there a correlation between student participation in CGU and their educational aspirations? Hypothesis I.1: Low income students that participated in CGU will demons trate higher aspirations to graduate from high school over time. Hypothesis I.2: Low income students that participated in CGU will demonstrate higher aspirations to earn a college degree over time. Hypothesis I.1 and 2 captures the idea that if students are exposed to information about college in a systemic way, their individual aspirations for post secondary education will be enhanced and that will show within survey results. In each of these questions the independent variable is exposure to CGU curricu lum and PCAs by the program participants versus compared to the control group not exposed to CGU curriculum and PCAs, and the dependent variables are their responses to survey questions. The rival hypothesis for Research Question I is that CGU does not imp rove low aspirations to earn a high school or college degree. Research Question II: Is there a correlation between CGU PCA mentoring and student aspirations to pursue postsecondary education? Hypothesis II.1: Low income students that part icipated in CGU will be more likely to pursue a postsecondary degree due to the strong relationship they developed with their PCA over time. Hypothesis II.1 addresses the idea that if students have a dedicated professional working with them on a regular b asis, the likelihood of them pursuing a postsecondary

PAGE 51

37 relationships with PCAs do not influence their pursuit of a postsecondary degree. Research Question III: Is t here a correlation between student participation in CGU and their knowledge and understanding of postsecondary pathways? Hypothesis III.1: Students aspirations to attend postsecondary education will increase over time. Hypothesis III.1 considers that if students are exposed to information about college in a way that is systematic with concepts introduced at specific times and revisited regularly, their aspirations to attend postsecondary education will increase as exhibited in their responses to survey q uestions. In each of these questions the independent variable is the implementation of the curriculum and the dependent variables are related to the specific questions. The rival hypothesis for Research Question III is that students do not gain hope in the ir ability to attend postsecondary education over time. Research Design This study is an evaluation perceptions of postsecondary education. The goal of the study was to utilize quantitative data as a metric to judge the value of a federally funded educational initiative. Gall et al. on program design though. Evaluations of program design tend to focus on identifying best practices rather than the study of their success (Garza, Barnett, Merchant, Soho & Smith, 2006). Gandara and Bial (2001) surveyed the range of K 12 intervention prog rams

PAGE 52

38 To best examine the CGU program and determine its general effectiveness, this uts and the activities implemented to address student success in an effort to learn if these elements were aligned with the goals of the program and to seek insight into whether they impacted the outputs, outcomes, and impacts of CGU. This logic model revi ew properly contextualized the findings of the research question as it described the roadmap used by the program to reach the desired results. Research Questions I, II, and III used a time series design that measured student responses to a survey as 8 th gr ade students compared to those same student responses as 11 th grade students. The questions highlighted for this study tried to measure two role that CGU played in In addition to survey data, the study utilized summative statistics gathered through the CGU to graduate high school and college Question I survey results were bolstered by high school graduation rates of CGU students who were in the program in 9 th grade and dropped out, most commonly because they moved to a non CGU high school, and the high school graduation rates of CGU stude nts who persisted in the program through high school graduation; CGU high school graduation rates compared to the graduation rates in their high schools the same year they completed high school; and the number of CGU college students at two year colleges, four year colleges, and selective four year colleges Colorado State Assessment Program. Question II includes college retention data from the

PAGE 53

39 ge semester to their second college semester compared to the retention accompl ish the same completion levels. Question III utilizes remediation statistics for CGU perceived understanding of postsecondary preparation compared to actual college readiness. SPSS was utilized with the 8 th grade surv ey and the 11 th grade survey There were a total of five q uestions parceled from the survey. Two of the questions addressed the students felt that CGU had in their pre co responses from 8 th grade and how they responded in 11 th grade was made. To clearly describe the survey questions that were examined within this study, those questions that are used to answer the research ques tion have been assigned a label, as detailed in the T able III.1 below. Table III.1 Survey Question Labels Used to Answer Research Questions I, II, and III Label Survey Question Number Research Question Number Survey Question High School Aspiration 2 I Do you think that you will graduate from high school? College Aspiration 3 I Do you think that you will go on for further education after you graduate from high school? High School Class 6 III Has anyone from your school or GEAR UP ever spoken to you about the following: Classes you need to take in

PAGE 54

40 Requirements high school to prepare for college? College Entrance Requirements 7 III Has anyone from your school or GEAR UP ever spoken to you about the following: College entrance r equirements? College Ability 10 II What is GEAR UP PCA doing for you? Rate how much you agree or disagree with the following statement: It will help me be able to go to college. Survey questions High School Aspiration and College Aspiration were groupe d together as a composite variable to answer Research Question I. Both questions centered on what students thought about their own abilities and aspirations. Students were asked in Question High School Aspiration if they believed they would graduate from high school and asked in Question College Aspiration if they thought that they would go to school beyond high school graduation. Students were given the choice of 5 responses: definitely probably not sure ), probably not and definitely not To answer Research Questions II and III, questions College Ability, High School Class Requirements, and College Entrance Requirements were used. Students were asked what role they believed CGU played in the m being able to go to college, what role CGU played in convincing them to go to college, what role the pre collegiate advisor played in their thoughts on college, and how much the students felt that CGU helped explain the financial aspects of attending col lege. For these questions student responses were: a lot some a little none In order for the key metric points mentioned above to be assessed, program

PAGE 55

41 implementation was essential. CGU began enrolling students as 7 th grade rs in 2006 2007. Students that were identified as free/reduced lunch eligible and/or students, who performed below grade level on State standardized tests, were selected to hear about CGU in a large setting at the partner school. In that meeting, CGU staff explained the benefits of the program. Students were encouraged to take enrollment forms home for parent signatures. The goal was to enroll 100 students per grade, per school. If enrollment did not reach 100 students, CGU staff was instructed to go and re cruit eligible students. Students were recruited in multiple ways. CGU solicited help from teachers, staff and even school administration to help fill the allotment. As the current administration was not in place at the beginning of this iteration of the CGU grant, there was no curriculum developed and no surveys implemented in the first year of this grant, surveys were developed and administered at beginning in the students 8 th grade year. It should be noted that there was a large drop in the number of st udents who took the survey as 8 th grade students and the number of students who took the survey as 11 th grade students. This can be attributable to the following factors: a large number of students had to be purged from the program due to the fact that th e prior administration allowed undocumented students to enroll in the program even though the grant was specific that it was only for United States citizens. This discrepancy be came clear when CGU began offering college courses and colleges needed social security numbers to enroll. A second facto r in the loss of kids was that three of the middle schools original to the grant were closed and students transferred to other middle schools. This changed matriculation patterns to high schools. All efforts were m ade to follow the students after the closing of the schools, but with limited resources, it proved difficult for

PAGE 56

42 CGU administrative staff to track those students. There were also students who moved out of the areas in which CGU was set to work. This popula tion is highly transient and students are in and out of schools. Finally, there were some students (less than 10) who were either removed from the program (for behavior issues) or opted out of the program. A list of all CGU students that started the progra m in the fall of 2006 would be used as the first point of reference. Data Collection To gather information to develop a logic model, CGU program documents including the grant application, an nual performance reviews, the CG U database, brochures and marke ting materials, financial records, training documents, Colorado statewide data, and information distributed to stakeholders were used. Background information was gathered through past and present executive directors of the program to learn how the componen ts of the program were established. This evaluation will establish whether inputs were utilized to effectively impact outputs. In each of the study questions, the independent variable is exposure to CGU curriculum and PCAs by the program participants vers us compared to the control group not exposed to CGU curriculum and PCAs. If the premise of financial difficulties and lack of awareness about college preventing college matriculation is to be proven true, a larger percentage of CGU students will have staye d in school, show more knowledge and have higher aspirations about post secondary education, and will take part in more college classes than students not enrolled in CGU. Over the course of the years of the administration of CGU very few things changed in the schools for which it partners. Certainly there is some turnover in school leaders, teachers and counselors. Almost all of

PAGE 57

43 the key District administration remained in place over the course of the grant. The dependent variable was the observed changes in student responses to the CGU survey starting in 8 th grade until the 11 th grade year. The overarching hypothesis is that students who participated in CGU would score higher on survey questions over time as measured by student responses on questions rela ted to understanding of the college admission process, financial aid opportunities, and the overall benefits of earning a postsecondary degree. The survey was designed by CGU administration staff, with help from an independent evaluator (a graduate stud ent from the University of Colorado). In the grant proposal to the United States Department of Education, an independent evaluator was provided for in the narrative. There were five specific questions required to be surveyed from the students in CGU and fi ve questions required to be surveyed from the parents through the duration of the grant. GU designed additional questions that supported program goals and desired outcomes. The student survey was administered online every January at all CGU partner school were sent home with students. All CGU students were required to complete the surveys. A control group of their peers were asked to complete the surveys as well. To try and ma tch up the students of the control group as much as possible with their peers, CGU ran the surveys through English classes at each partner school that were not considered honors courses. It should be remembered that much of the cohort of CGU students being surveyed were part of CGU because they did poorly on State standardized tests.

PAGE 58

44 Validity There are a few reasons to have concerns of internal validity within this study, however none of these reasons would be different from general concerns of studying s responses over time as these are individuals who typically face difficult situations in their community and family, including drug use, violence, early pregnancy, unemploymen t, and homelessness. Study participants could have faced any one of more of these factors throughout their time within GEAR UP. Repeated testing could have led to student bias. Participants may have been conditioned through program lessons to answer that they do understand the process to attend and graduate from postsecondary school and that they plan to do so even if this was not the case. However, statistics regarding student entry into postsecondary education gathered by CGU seem to indicate students w ere sincere in their answers. While student maturation could have impacted their answer s to survey questions, this is mitigated by the fact that all of the study participants most likely faced the same changes. The demographics of each school student pop ulation do not change much from year to year. Certainly, there is a chance that there are students that might be considered outliers, but it is probably safe to say that there is no real threat to internal validity. Students were allowed into the program based on 1) their low income status based on their eligibility for the National School Lunch Program, and 2) t heir low performance on state wide academic standardized tests (CSAP). Previous research has proven that students from lower socioeconomic backgr ounds, regardless of race, ethnicity, or desire to attend college, are unlikely to enroll in postsecondary education and persist to

PAGE 59

45 graduation. This background makes it clear selection bias is not an issue. Even if the students do exhibit traits or desires that make them more interested in college, data proves their interests do not lead to enrollment or degree completion. For future studies, one way to avoid any selection bias would be to survey students before they were selected for the program and only take those that fell at or below a median response range to varying college going questions. As for threats to external validity, they do exist. The Novelty Effect speaks to the tly than they would if the experience was one in which they had before. The goal of the CGU program was to remove financial concerns as a barrier and enlist the aid of caring adults providing specific information about college. It could be assumed that a s tudent that chooses to enroll in CGU is already properly motivated to attend college. An additional threat could be the change in N from 8 th to 11 th grade as the number of participant survey takers drops by over half. This number was due to a large numbe r of undocumented students who were initially selected into the program having to be removed from the program due to federal regulations regarding program participation and residency. These students could have influenced changes in population responses and resulting statistics from 8 th to 11 th grade. However, their responses to the 8 th grade survey would have been provided in an environment where they had been informed by a CGU advisor that postsecondary enrollment was possible, and where they were within a school that was mandated to pay for their education up to high school graduation. These elements would have likely mainstreamed their responses with documented participants that completed the survey. Additionally, they did not complete the 11 th grade surv ey, and

PAGE 60

46 would not have had a negative or positive impact on the results. Data Analysis First, a review of programmatic documents including the grant application, an nual performance reviews, the CG U database, brochures and marketing materials, financial re cords, training documents, Colorado statewide data, and information distributed to stakeholders illuminated the varying elements of the program including resource inputs, activities, outputs, outcomes, and impacts. Analysis of this information show ed that the elements available to the program were properly implemented to effectively reach the desired goals. This was determined if implementation occurred according to the design framework. This set a baseline of programmatic understanding within this disserta tion, and served as the initial point of determination as to whether the program was properly placed to reach the results it sought. Information regarding graduation rates, retention rates, remedial needs, and CSAP rates were analyzed using Microsoft Exce l to create participant percentages compared to full class percentages.

PAGE 61

47 CHAPTER 4 FINDINGS Introduction This quantitative research study discovered how CGU was designed and implemented in an effort to improve student knowledge of collegiate pathwa ys, increase their postsecondary aspirations, and utilize pre collegiate advisors (PCAs) to lead students to pursue postsecondary degrees. Three primary research questions were established and examined. Table IV.1 shows the research questions and the sampl e size for each question. Table IV.1 Research Questions and Sample Sizes Research Question Survey Question 8th Grade Participants 8th Grade Control 11th Grade Participants 11th Grade Control I Do you think that you will graduate from high school? 1387 90 1182 1396 I Do you think that you will go on for further education after you leave high school? 1385 90 1182 1394 II What is your GEAR UP PCA doing for you? Rate how much you agree or disagree with the following statement: She/he will help me be ab le to go to college. 1380 88 1172 1372 III Has anyone from your school or GEAR UP ever spoken to you about Classes 1389 90 1182 1388

PAGE 62

48 you need to take in high school to prepare for college? III Has anyone from your school or GEAR UP ever spoken to you abo ut College entrance requirements? 1389 90 1176 1380 Research Question I found there is a statistically significant difference in participants aspirations to attend college from their 8 th grade year to their 11 th grade year, and aspirations among particip ants were greater than those of the control group. Research Question II found participants did build effective relationships with their school PCA. Research Question III found there was a significant difference between CGU participants and the control grou p as well as between CGU participants in 8 th and then 11 th grade regarding their college aspirations. To clearly describe the survey questions that will be examined within this study, those questions that are used to answer the research question have been assigned a label. Please review Table 1 in Chapter 3. CGU Logic Model Design and Implementation A logic model was used to better understand how the program utilized resources and implemented activities, and how those efforts impacted outcomes. Figure IV. 2 represents the numerous elements that go into the planning, implementation and execution of CGU, as well as the intended results. Each of these elements and their relationships are des cribed in the following section.

PAGE 63

49 Figure IV. 1 Logic Model Flow Chart

PAGE 64

50 Logic Model Design Resource inputs. Financial resources The CGU grant received 3.5 million dollars per year from the United States Department of Education from 2005 2011, for a total of 21 million dollars. Of the 3.5 million dollars annually, 1.75 mi llion each year went directly into a trust fund for student scholarships. The other 1.75 million each year went to operational costs. Obviously, the money supplied from the Federal Government is essential to the program. Being able to offer student scholar ships is an important motivating factor for students. Having money for operational costs is just as essential. For CGU being able to utilize those dollars as they saw fit aided their efforts. However, more guidance and guidelines from the United States Dep artment of Education are needed for best implementation across the country. Organizational resources. The CGU staff consisted of 13 full time PCAs, four part time pre collegiate advisors, an Executive Director, Program Director, a Director of College Pat hways, a Budget Director, a part time Data Resource Manager, and a Financial Aid and Scholarships Director. Total staff equated to 18 full time and five part time staff. All of the roles initiated by the CGU team proved to be critical to the success of the program. The Executive Director set the overall vision for the program and collaborated with school districts and colleges across the State to implement strategies. The Program Director developed the curriculum that PCAs utilized with their students and a lso managed PCA staff. The Director of College Pathways worked with the

PAGE 65

51 individual schools and their partner community colleges and 4 year schools to implement dual enrollment courses, CLEP testing, and remediation courses. The Budget Director was responsi ble for the day to day operations but was also responsible in creating enough rollover money for the program to go one extra year. This helped 1 st year college attenders dramatically by providing on campus support by PCAs. The part time Data Resource Mana ger developed the data base and helped with all key reporting. The Scholarship Director maintained all records of scholarships but also provided all trainings on financial aid to the PCAs. Finally, the PCAs were most important in the motivation of their st udents. Community resources. CGU partnered with eight school districts across Colorado, 12 high schools, 20 community colleges and universities and two career and technical schools. All efforts were made to provide for geographic diversity. Private educa tional services. CGU partnered with ACT to offer the EXPLORE and the PLAN in some of the partner schools. All CGU students take the ACT exam in the 11 th grade. CGU partnered with College Board to offer the College Level Exam Placement test (CLEP). Student activities. Academic counseling Students met with PCAs at least twice a month from the time they were in 7 th grade through their senior year in high school. Students who were struggling, either personally or academically, were met with more often. The n umber of the visits, the date of the visit, and the reason for the visit were catalogued in the organizational data base.

PAGE 66

52 Through reviews of programmatic planning documents, CGU utilized grant dollars to pay for PCAs, tutoring, textbook needs, remediation coursework, dual enrollment, transcript reviews to assure students were completing rigorous coursework, and communication with teachers. College visits Beginning as early as their sophomore year in high school, students began taking college visits. Initi ally they were sponsored by a partnership between CGU, the school district and/or high school, and the college or university and entail a large number of students. As students got older, there was an increase in the amount of specific visits. Those were us ually sponsored by CGU and the specific postsecondary institution. All visits were catalogued in the CGU organizational data base. CGU documents showed the program utilized financial resources to help fund college visits. This included money for school and charter busses, dorm room fees for overnight stays, money for food, and college memorabilia. CGU used their organizational resources such as staff to schedule those college visits and to provide chaperones for those visits. As for community resources, CGU partnered with the high schools and their districts to schedule busses, and to include students who were interested in the particular college or university but not part of CGU, share costs, and provide school personnel as chaperones. Colleges and universi ties were utilized to help schedule visits and campus tours, help share in the cost of food and college memorabilia, and to provide housing in the case of overnight stays. Private educational services and community partners were utilized to provide volunte ers, as well as incentives for campus visits.

PAGE 67

53 PCA activities. Trainings/retreats. Pre Collegiate advisors received trainings in many ways throughout the grant. There were two mandatory national meetings per year, in which GEAR UP programs from across t he country are required to send representatives. As part of the state implementation, there were two statewide meeting each year that are typically held in August and in January. Operational dollars were utilized to provide transportation to the sites, foo d and lodging. The CGU Program Director then conducted quarterly meetings with a smaller group of advisors, typically usually grouped by geographic area. In addition, specific trainings were posted on the CGU website in the form of webinars. This enabled a dvisors to review training at their convenience and to review it as often as needed. The organizational resources of CGU worked together in those retreats to disseminate information, review programmatic goals, review data, share best practices and to strat egize future challenges. Monthly meetings were conducted to review data and strategies between CGU administrators and PCAs in a one on one setting. CGU partnered with community organizations to provide meeting space, speakers and informational material. Ad ditionally, school districts often provided pertinent data to guide conversation and decisions. Private educational services information was disseminated in the form of ACT preparation information and CLEP test information, while representatives from these organizations were present for trainings to fully equip PCAs to utilize their services and products.

PAGE 68

54 These trainings were crucial in a couple of ways. First, not all of the CGU PCAs were as knowledgeable about college pathway information as they might hav e needed to be. Providing knowledge through these trainings increased their capacity. Trainings were also important because it helped CGU have a uniformed and systemic way of disseminating college pathway information. Staff bonuses/incentives PCAs were eligible to receive bonuses based on student performance outcomes. Measures included: 90 % student program retention; 90% success rate for high school students taking college courses (success is equated t of students receivin g more than one scholarship award; 90% participation in summer programs including community service, job s hadows, and internships; and 90% success in summer courses (high school or college). CGU utilized financial resources through operational dollars to pay for those incentives. CGU used organizational resources in the form of administration monitoring metrics for bonuses and by providing training and support to provide PCAs with best practices and strategies to achieve those incentives. Community resourc es were utilized as high schools provided some of the necessities to ensure successful implementation of the program. High schools provided office space, computers, telephones and access to student data information. Bonuses and incentives served two purpos es. Because there was a state wide salary freeze it was the only avenue CGU administration had to fiscally acknowledge the great work that PCAs were doing. The bonus and incentive program also rewarded particular PCAs who were getting the best results.

PAGE 69

55 Ann ual conferences In reviewing the GEAR UP grant as set forth by the United States Department of Education, there were two national meetings each year. It is a requirement of the grant that each state provide at least six representatives to each conferenc e. CGU rotated the representation in a way that all PCAs had an opportunity to attend one national conference every three years which was used as a staff development opportunity. The financial resources to accomplish these trainings were covered by DOE and had to be included in each state program budget. Organizational resources were committed to attend. CGU typically sent teams of 6, made up of administration and PCAs to these conferences. Private educational services were provided by the National Council for Community and Education Partnerships (NCCEP). NCCEP organized the conferences and they provided break out sessions for a wide array of best practices. Professional development Pre collegiate advisors and administration staff were given work time (and on occasion financial aid) to attend related professional training opportunities. Financial resources for professional development had to be designated by the individual grants. In reviewing records, CGU handled a large amount of their professional devel opment in house. CGU did not invest a lot of money in outside professional development opportunities. Most of the professional development was handled by CGU administration staff. A high percentage of organizational resources were designated to training an d development of PCAs. Beyond the two organizational retreats, there were monthly team meetings including a few PCAs linked together by geographic location and members of

PAGE 70

56 the administration team, and weekly phone calls between administration and individual PCAs. There was also a weekly review of PCA entered data that was conducted by CGU administration. Meetings were held at school sites and schools commit office space to these meetings. CGU used a robust, multifaceted to provide reliable and comprehensive information for nearly every aspect of the program. Information from the database help ed influence outcomes in secondary and post secondary school s by giving the pre coll egiate preparation. A thorough plan was developed to ensure that data was assess ed program outcomes that included progress; attendan ce; testing and assessment scores; grade point average (GPA) by subject; enrollment in rigorous courses required for graduation; concurrent enrollment kind contributions; and services p rovided to students, parents, and teachers. In addition, student records captured in the CGU database were matched with application, enrollment, remediation, and financial aid files collected by CDHE and state K 12 academic records collected by CDHE. Fro m all of this data, the program was able to do thorough data mining to assess the program and intervene early when students were not meeting requirements or are falling behind. The database has a customizable reporting function that allow ed users to pull reports querying any of the types of data that were collected. Some of these reports include d tracking mentoring services including academic inventions related to academic

PAGE 71

57 progress and attendance; tracking monthly post secondary readiness workshops to en sure all students are receiving the same preparation in a timely manner; tracking PCAs task lists such as the Senior Checklist a comprehensive list that ensure all students who apply and commit to a college register for courses; complete orientation and o ther college enrollment steps; and allowing for the efficient tracking of the various stages of student activities until completion, such as completing the FAFSA or 4caster for juniors and seniors. The database also has an online student college portfoli o function which provide d the student with the ability to interact with the database and review their own specific academic, financial aid, and college preparation information, as well as provides a place for them to save documents for college applications such as their college essays, transcripts, work experience and extra curricular activities. Since most of the CGU students d id not have their own computers at home, providing them with the ability to access this portfolio on any computer with an internet connection ensure d that these documents were not lost and were submitted on time. and the management team ha d access to all of the student portfolios and utilize the reporting features to ensure all students were completing assignments and steps fo r college preparation and enrollment. The CGU College Portfolio allow ed students to track rigorous coursework such as the Colorado Higher Education Admission Requirements (HEAR), research and compare colleges, write a personal statement, and search and ap ply for scholarships along with other activities that relate to college preparation and admission.

PAGE 72

58 Outputs. Student knowledge of college requirements All students were given a survey in the spring of each year to determine overall knowledge of college requirements. A control group of GEAR UP students peers were also administered a survey to provide a comparison. Comparisons were drawn between CGU schools to help gauge advisor effectiveness. Academic counseling was administered to the students at least twice a month for the duration of their time in CGU. In one of those meetings, students met with PCAs in groups to discuss a college curriculum package that was developed by CGU administration. Typically, concepts were introduced in the early grades (7 8 and 9) and are built upon and made more student specific, in later grades (10 11 and 12). Students began to go on college visits their sophomore and junior years of high school. In discussions with the CGU administration team, it is their belief that co llege visits too student becomes older. Student knowledge of financial aid All students were given a survey in the spring of each year to determine overall knowledge of college requirements. A control group of GEAR UP students peers, were also administered the survey to provide a comparison. Comparisons were drawn between GEAR UP schools to help gauge advisor effectiveness. Knowledge of financial aid is critical to stu dent matriculation and success at the postsecondary level. Students were introduced to concepts early in their CGU participation, and those messages were

PAGE 73

59 revisited often. All students were asked to complete a FAFSA 4Caster report as juniors and to fill out a FAFSA form as seniors. Improve grades and test scores Student GPAs and test scores were tracked using the organization data base. High school grades, CSAP scores, and college grades were all utilized to help determine program effectiveness. CGU staff and administration attended biannual GEAR UP conferences to stay current on best practices for improving grades and attendance. In discussions with CGU administration and staff, they believe constantly and consistently talking about college and college pat hways is the best motivation for students to attend school regularly and work hard to achieve. CGU PCAs attended as many professional development trainings as possible. They were involved in weekly and monthly training with CGU administration. They also ha d many opportunities to participate in school and district professional development workshops. It should be noted here that CGU could only have an indirect impact on student grades and test scores. PCAs had access to student school data and were able to he lp students identify problem areas in specific courses early in semesters. They were also there to help advocate on the students behalf and to teach students how to self advocate. PCAs did not teach any courses and thus were not able to directly impact gra des or test scores. That is the role of the schools. Motivation to attend college All students were given a survey in the spring of each year to determine overall knowledge of college requirements. A control group of GEAR UP students peers, were also ad ministered a survey to provide a comparison. Comparisons were also drawn

PAGE 74

60 between GEAR UP schools to help gauge advisor effectiveness. CGU administration and PCAs believe it was their primary role to motivate students to attend college. Students were made a ware of postsecondary pathway information early and often in their GEAR UP careers. CGU s administration felt the more that students hear the message about college; the more students will assume that is the expectation. As important as it was for PCAs to d isseminate the proper college pathway information, motivating their students was perhaps the most important role that PCAs played in the CGU program. When working with traditionally underserved students, knowing the information is just a part of helping st udents achieve success. What is so difficult is constantly and consistently pushing students to believe in themselves that they are capable of attending and being successful in college. Outcomes. Increase postsecondary aspirations, awareness, and knowledge All students were given a survey in the spring of each year to determine overall knowledge of college requirements. A control group of GEAR UP students peers, were also administered the survey to provide a comparison. Comparisons were drawn between GEAR time a year to CGU students and a group of their peers to determine if there was a significant difference in student aspirations. Those findings are detailed in research questio ns II and III. If other pre collegiate programs do not begin to take a more active role in this area, the achievement gap is going to continue to widen and millions of dollars will be

PAGE 75

61 on approach through their PCA s should be replicated in other pre collegiate programs throughout the country. Enhanced academic achievement and development Student grades were monitored each semester and registered in the organization data base. Particular attention was paid to colle ge course grades and matriculation through college remedial coursework. Again, it should be noted that CGU had only an indirect impact on the success of their students in these areas. Schools, teachers, and counselors had direct impact. Increased postsecon dary participation Completed student FAFSA forms served as a strong indicator, but all student contact information was logged and follow ups were done with the individual students as well as all participating partner colleges and universities (through fi nancial aid offices and census data). The bottom line for any pre collegiate program is right here. Did their students percentage for students of similar populations (stu dents from low socioeconomic backgrounds) attending college in 2011 as first time, full time students was 42%. Impacts. Short term benefits to students Significant dollars were utilized to pay for GEAR UP students to take college classes, CLEP tests, and developmental education courses. Full time PCAs were assigned to students to help them become familiar with college concepts, advocate for them, teach them how to advocate for themselves, and guide them through middle school, high school

PAGE 76

62 and through their pathway information is equally critical. Too often students are into their senior year of high school before they realize the value of a good grade point average or the importance of a standardized test such as the ACT. Interaction with the PCAs on a regular basis helps mitigate that concern. across the educational system. In discussions with CGU staff and ad ministration, the unknown benefit of financial aid is one of the biggest hurdles that students from at risk populations face. CGU used their case management strategy; a caring adult providing a focused curriculum, and the constant and consistent emphasis o n college to motivate students to work hard in school. GEAR UP measured student graduation rates, percentage of students who completed the FAFSA form, postsecondary matriculation rates, remediation percentages and college retention to measure effectiveness CGU relied on quantitative data to prove that their methods as a pre collegiate program showed that students who participate in CGU have a greater motivation to achieve postsecondary success, and know more about postsecondary pathways. CGU looked to high school graduation rates, college matriculation rates and remediation participation rates to determine their success. Long term benefits. Considering the first educational hurdle for at risk students, high school graduation, CGU had an impact on participa economy. The U.S. Department of Revenue estimates that the total tax payments for any high school graduate is around $139,100 for a lifetime, that they save the federal

PAGE 77

63 government $40,500 on public health exp enditures, do not create $26,600 in law enforcement costs, and do not utilize $3,000 in Welfare. The average aggregate lifetime public economic benefit is $209,100. Considering this is for just one individual, CGU was not only important to the lives of tho se students participating in the program, it helped provide an economic efficiency tool for the federal government as well. In the education sector it is typical to have personnel costs account for well over t personnel costs, which included salaries, wages, and benefits, only accounted for approximately 38% of the total six year grant total of 21 million dollars. This means a very high proportion of federal dollars went directly to benefiting student partici pants. CG cost was reasonable relative to the benefits that the students received: including completing remediation needs while in high school, earning college credit while in high school, earning a high school diploma, enrolling in college and increasi ng their chance of graduating from college. College degree holders earn nearly twice as much personal income as those with only a high school diploma, they become and remain employed, and are better able to adapt to the ever changing workplace ( Retrieved f rom http://highered.colorado.gov/Publications/Studies/2007/200712 ImpactofHE.pdf, 2007) The CGU model yielded a very high return on investment. The National Association of Colleges and Employers found that the current 2011 college graduate makes an ann ual salary of $50,462 (Retrieved f rom http://www.naceweb.org/research/salary survey/ ? referral =research&menuID=71, 2011). The project recruited 1,200 students for four years starting in 7 th grade, totaling 4800 students. Over the seven year life of th e grant, there was an estimated 26,400 duplicated

PAGE 78

64 enrollments in the program throughout the grant. This total number of enrollments divided by the operating budget of 10.5 million dollars equals on average a cost of $398 per enrollment per year. If the 1 0.5 million dollar scholarship component is included in the calculation the average cost of an enrollment per year doubles to $796. Studies have shown that the taxpayer benefit from high state and federal income taxes paid on the higher salaries earned by college graduates varies from $60,000 to $150,000 in additional income taxes paid over the work life of the graduate, depending on the selectivity of the college attended. This easily surpasses the total cost per student of the CGU program. Not only wil l college graduates pay more in taxes but they will also be less likely to rely on government services and support programs throughout their life. Logic Model and Program Implementation In every school district, the potential director of CGU began meeti ng with school districts involved in the grant before it was learned of Colorado won the grant. This was necessary to assure the program would be able to recruit students immediately after a federal determination on CGU s application was communicated. Thes e meetings included CGU, the district superintendent the school principals, and other representatives invited by the district. In these meetings, CGU communicated all aspects of the program to all parties. Also before the receipt of a federal determinatio n, CGU invited applications for employment and interviewed potential advisors. The requirements included previous experience working in schools, previous advising experience, a degree, previous experience with students ages 12 20, and previous e xperience with underrepresented and low income students. While the application process and requirements were the same for all advisors, those selected had varying degrees of

PAGE 79

65 experience as it was easier to find highly qualified candidates to work in the mor e populated areas of the state than it was in the rural areas of the state. However, every candidate met the minimum requirements for employment. CGU began receiving federal grant dollars in September of 2006. Participating schools were informed the gran t would move forward within the week. Advisors were informed CGU had received the grant, and were brought to the Colorado Department of Higher Education for a week long training. In this training, all aspects of the program were described including student services, communication with school representatives, database training, review of reporting requirements, and conduct expectations. Other activities included the distribution of phones for all advisors, computers for those who would not receive them from t he district, and team building activities. Advisors began their work in schools the next week Schools provided office space, database access, classroom access, computer access, as well as all other tools or resources provided to the school counselors. In some schools, acquisition of computers, office space, or database access was challenging either due to a lack of district resources or Information Technology right of usage policies. Some districts were hesitant to allow access to student records. When it was clear any resource was not coming from the district, CGU provided them. Regarding student record information, within two months all districts had provided access to all advisors. Student recruitment began in October and finished in December of 2006, w hile student services began in January of the same year.

PAGE 80

66 Research Question I: CGU Impact on Student Educational Aspirations. The results from survey question s on High School and College Aspiration demonstrate that participation in the CGU program does increase the likelihood that participating students will understand the value of a post secondary degree. This is shown in the increase of students within the program that state they will seek out additional education upon completion of high school. It i s also demonstrated that CGU participants are more likely to pursue additional education compared to the control group. CGU vs. Control It was important to draw references between an experimental CGU group and a control group of their peers. CGU particip ant numbers decreased from 8 th to 11 th grade due to matriculation out of GEAR UP schools or leaving the program, and control group participants increased due to greater recruitment. Participants in CGU had a higher average on their responses in comparison to the control group for survey questions on High School Aspiration and College Aspiration. Participants in CGU were more confident they would complete high school and pursue additional education than their peer counterparts in the control group. To br graduate from high school, Figure IV. 1 shows how after one year of services CGU students were more likely to answer affirmatively that they would graduate from high school. Through mon thly meetings with CGU PCAs, participants learned the importance of high school graduation to postsecondary success, and continued to aspire to high school graduation at a higher rate than their peers. In total, both participants and CGU student peers aspi red to high school graduation at a higher rate each year, with

PAGE 81

67 roughly 10% more CGU participants expecting to graduate than their peers each year the survey was taken. Figure IV. 2 CGU versus Control Group: Do You Think That You Will Graduate From High Sch ool? High School Aspiration and College Aspiration A survey was utilized t o understand the impact tha t CGU had on student aspiration. Students were first surveyed during the ir 8 th grade. The survey consisted of 51 questions Specific questions on their high school aspirations, their college aspirations, and their expectations of level of education att ainment were chosen to address Research Q uestion I To compare these answers to students who were not part of the CGU program, the survey was also administered to a control group of their peers. To best match up the control group with the experimental group, students from the control group were chosen by the level of their English course. As the CGU students were mostly identifie d by

PAGE 82

68 control group chosen were in mainstream English classes. The same survey questions w ere then administered to the same participants and control group students i n 11 th grade. The same steps to identify the control group were taken. A comparison of the two groups in the 8 th grade and then again in the 11 th grade is below. For survey q uestion High School Aspiration graduate from High Table IV: 2 Descriptive Statistics Research Question I, Survey Question High Schoo l Aspiration N Mean Std. Deviation Do you think that you will graduate from high school? Control 90 4.64 .641 Gear Up 1387 4.61 .632 Do you think that you will graduate from high school? Control 1396 4.73 .560 Gear Up 1182 4.83 .458

PAGE 83

69 Table IV: 3 Independent Samples T Test Research Q uestion I, Survey Question High School Aspiration Levene's Test for Equality of Variances t test for Equality of Means 95% Confidence Interval of the Difference F Sig. t df Sig. (2 tailed) Mean Diff erence Std. Error Difference Lower Upper 8 th Grade Do you think that you will graduate from high school? Equal variances assumed .054 .817 .554 1475 .580 .038 .069 .097 .173 Equal variances not assumed .547 100.528 .586 .038 .070 .100 .176 11 th Grade Do you think that you will graduate from high school? Equal variances assumed 80.161 .000 4.964 2576 .000 .101 .020 .141 .061 Equal variances not assumed 5.046 2573.048 .000 .101 .020 .141 .062

PAGE 84

70 An independent samples t test was con ducted to compare the attitudes of high school students participating in CGU and a control group of high school students regarding whether they think they will graduate from high school. This attitude question was asked at two different time periods, first when the students were in 8 th grade and then when the students were in 11 th grade. There was no significant difference in the attitude scores between CGU students (M=4.61, SD=.632) and the control group (M=4.64, SD=.632) on the 8 th grade survey; t=.554, p =.580 In the second survey period there was a significant difference between the attitude scores of CGU students (M=4.83, SD=.458) and the control group (M=4.73, SD=.560) on the 11 th grade survey; t= 5.046, p=.000. High School Aspiration High school aspir ation 8 th grade. Table IV. 4 Model Summary Research Question I, Survey Question High School Aspiration 8 th Grade Model R R Square Adjusted R Square Std. Error of the Estimate 1 .152 a .023 .019 .314 Table IV. 5 ANOVA Research Question I, Survey Que stion High School Aspiration 8 th Grade Model Sum of Squares df Mean Square F Sig. 1 Regression 13.581 22 .617 6.273 .000 a Residual 576.228 5855 .098 Total 589.809 5877

PAGE 85

71 Table IV. 6 Coefficients Research Question I, Survey Question High Sc hool Aspiration 8 th Grade Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics B Std. Error Beta Tolerance VIF 1 (Constant) 4.659 .024 196.342 .000 GroupCode .003 .014 .004 .190 .849 .343 2.917 gender .03 3 .011 .038 2.959 .003 .988 1.012 American Indian .103 .048 .028 2.128 .033 .946 1.057 Asian .060 .039 .021 1.538 .124 .902 1.108 African American .038 .024 .023 1.568 .117 .747 1.339 Hispanic .016 .015 .024 1.046 .296 .320 3.127 Native Hawaii an .012 .085 .002 .145 .885 .983 1.017 Not Reported .093 .056 .022 1.662 .096 .968 1.033 White .024 .018 .023 1.341 .180 .579 1.726 Freed Middle .052 .061 .012 .852 .394 .842 1.187 Lamar HS .038 .315 .002 .121 .903 .994 1.006 Lamar Middl e .066 .032 .038 2.074 .038 .499 2.005 Pueblo East HS .162 .159 .013 1.022 .307 .979 1.021

PAGE 86

72 AWCPA .036 .033 .020 1.097 .273 .496 2.014 Grand Mesa Middle .064 .037 .029 1.731 .084 .597 1.676 Franklin Middle .195 .034 .098 5.662 .000 .558 1.791 Ortega Middle .075 .035 .037 2.150 .032 .568 1.762 Colorado Springs East Middle 2.476 1.177 .975 2.104 .035 .001 1287.548 Martin Luther King .070 .033 .246 2.097 .036 .012 82.594 Kunsmiller Middle .167 .032 .590 5.283 .000 .013 74.6 43 Mt. Garfield Middle .082 .069 .292 1.193 .233 .003 359.201 Jefferson Middle .035 .058 .123 .595 .552 .004 256.280

PAGE 87

73 A linear regression was performed in SPSS to determine if the following independent or predictor variables: gender, ethnicity, m iddle school, and participant in CGU or not could be used to estimate a student's attitude regarding graduating from high school. Using these predictor variables a significant model emerged (F=6.27, p =. 000). Despite the significance of the predictor model the Adjusted R square = .019, suggests low predictability strength and the variables in the model only account for under 5% of the variance in the survey question answer. Several independent variables in the model were significant including: gender Beta =. 0 38 p=.0 33 American Indian Beta = .0 33 p=.0 28 Lamar Middle School Beta = .038 p=.0 38 Franklin Middle School Beta = 098 p=.00 0 Ortega Middle school Beta = .037 p=.0 32, East Middle School Beta = .975, p= .035, and Martin Luther King Middle School Beta .246, p= .036. These significant independent variables are contributing to a portion of the variance in the regression model. The Beta coefficients display the unique effect of an independent variable on the dependent variable. Specifically and for example, for the gender variable the expected change in attitudes towards graduating high school is .038 change going from males to females. Or a student attending Martin Luth er King Middle School has a .246 one unit change going from this school compared to other middle schools, holding the other covariates constant. These results suggest that females are more likely to answer slightly higher on the survey question regarding attitudes about graduating high school and that there are difference among ethnicities and middle schools and their effect on predicting the survey responses.

PAGE 88

74 Collinearity diagnostics were run to determine if there were high correlations amongst the pred ictor variables. A tolerance level of .01 was used. Applying this criterion three middle school s fall below the .01 tolerance level therefore suggest ing there might be multicollinearity issue with the middle school variable. The collinearity between some of the middle schools makes it challenging to draw inferences about the relative contribution of each predictor variable to the overall success of the model. However, the middle school variable was not excluded from the model because it has practical impli cations All of the three middle schools that could have a multicollinearity issue were associated with Colorado Springs. This could indicate an issue with the CGU PCA employed at the time. It should be noted that employee was terminated for other reasons at the end of that school year. High school aspiration 11 th grade. Table IV. 7 Model Summary Research Question I, Survey Question High School Aspiration 11 th Grade Model R R Square Adjusted R Square Std. Error of the Estimate 1 .193 a .037 .027 .498 Table IV. 8 ANOVA Research Question I, Survey Question High School Aspiration 11 th Grade Model Sum of Squares df Mean Square F Sig. 1 Regression 17.375 20 .869 3.497 .000 a Residual 447.661 1802 .248 Total 465.036 1822

PAGE 89

75 Table IV. 9 Coeffici ents Research Question I, Survey Question High School Aspiration 11 th Grade Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics B Std. Error Beta Tolerance VIF 1 (Constant) 4.661 .042 112.230 .000 GroupC ode .140 .026 .132 5.480 .000 .916 1.092 gender .046 .024 .046 1.960 .050 .977 1.023 American Indian .018 .100 .004 .181 .857 .977 1.024 Asian .142 .076 .044 1.870 .062 .955 1.047 African American .144 .050 .072 2.865 .004 .842 1.188 Native Ha waiian .172 .152 .026 1.130 .259 .978 1.022 Not Reported .254 .106 .056 2.396 .017 .974 1.027 White .130 .033 .105 3.913 .000 .738 1.355 Lamar HS .079 .051 .050 1.550 .121 .518 1.931 AWCPA .135 .117 .028 1.151 .250 .913 1.096 Martin Luther King .005 .058 .002 .080 .936 .616 1.623 Aurora Central High .009 .051 .006 .179 .858 .560 1.785

PAGE 90

76 Grand Junction Central HS .017 .056 .009 .295 .768 .551 1.815 Northridge HS .066 .052 .039 1.276 .202 .577 1.734 Greeley Central HS .015 .094 .004 .163 .871 .865 1.157 Alamosa HS .032 .054 .018 .606 .545 .592 1.689 Wasson HS .023 .063 .010 .363 .716 .657 1.522 North HS .058 .055 .031 1.067 .286 .617 1.620 Montbello HS .021 .060 .010 .353 .724 .628 1.593 Abraham Lincoln HS .04 8 .051 .029 .939 .348 .568 1.760

PAGE 91

77 A linear regression was performed to determine if the following independent or predictor variables: gender, ethnicity, high school, and participant in Gear UP or not could be used to estimate a student's attitude regard ing graduating from high school. Using these predictor variables a significant model emerged (F=3.50, p = 000). Despite the significance of the predictor model the Adjusted R square = .027, suggests low predictability strength and the variables in the model only account for under 5% of the variance in the survey question answer. Several independent variables in the model were significant including: group code Beta =. 132 p=.0 00 gender Beta = .0 46 p=.0 50 African American Beta = .072, p=.004 and White Beta = .105 p=.000. These significant independent variables are contributing to a portion of the variance in the regression model. Holding all the other independent variables constant in the model, a typical CGU student is likely to answer .132 points more posi tively on the five point Likert scale than a student not in the CGU program regarding the likelihood they will graduate high sc hool. This means if the mean control group score for this question is 4.73 CGU students would on average answer 4.862 Holding a ll the other independent variables constant in the model, females compared to males tend to answer about .05 points more positively on the Likert scale regarding the likelihood they will graduate high school. African American students (.07) and Caucasian s tudents (.11) answer more positively than their Hispanic classmates. The base level for all the regressions was chosen by picking the largest of the groups, in all cases that was Hispanics for the ethnic groups, Skinner Middle School and Aurora Central Hig h School. Those students in CGU answered more positively than their peers would indicate that over the course of time that being part of CGU increased their expectations of high school graduation at a higher rate

PAGE 92

78 than their peers. Another interesting thing to note is that females again answered more positively than their male counterparts. It is difficult to generalize as to why African American and White students had a more positive expectation than their peers. Essentially what the regression shows is th at with other variables in the regression controlled for in the regression that the students in CGU answered more positively in the 11 th grade than their peers as to whether or not they thought they would graduate high school. This proved to be true for th e students who remained in the CGU program. It is difficult to draw any other strong conclusions from the variables that tested as statistically significant. Collinearity diagnostics were run to determine if there were high correlations amongst the predict or variables. A tolerance level of .01 was used. Applying this criterion no variable falls below the .01 tolerance level therefore suggests there is not a multicollinearity issue with any of the variables.

PAGE 93

79 GEAR UP graduation rates. In addition to 8 th and 11 th grade student aspirations to complete high school, data highlights the graduation rate of students who began in the GEAR UP program in 9 th grade and those that contin ued through high school graduation. This table moves beyond aspirations to actual graduation statistics. Table IV: 10 CGU 9 th Grade and 12 th Grade Graduation Rates Cohort Grade # Cohort # Graduated Grad % 9 569 350 61.5% 12 305 275 90.2% Graduation st atistics were tabulated for CGU participants who completed the program and those who did not. Graduation rates for those that did not continue within GEAR UP were gathered through statewide data provided by school districts. Of those that began in the prog ram but did not finish, 61.5% graduated from high school. Of those students that completed the program, 90.2% graduated from high school. This is compared to a Colorado state average of 75% This shows that students who were willing to both learn a bout the college going process and stay at the CGU partner high school were more likely to graduate high school than those who did not complete the program It is important to remember here that the largest percentages of these students were scoring lower in stand ardized aptitude tests than their peers at point of entry. GEAR UP curriculum, college application assistance, concurrent enrollment options, and remedial

PAGE 94

80 likelihood to gradua te high school. Table I V.10 shows the graduation rate of students in CGU as well as their schools graduation rate for the same year they completed school. Cohorts in two of 12 CGU schools exceed the state graduation rate, and only four of 12 overall scho ols exceed the state graduation rate, proving the schools CGU works within are in the bottom half of Colorado schools regarding graduation rate. Eight of 12 CGU cohorts underperformed their school classmates, which possibly highlighted how CGU selected stu dents into the program who were struggling to meet graduation expectations. The points of interest here are some of the massive disparities in a few of the schools in graduation rates. In discussions with CGU administration, the disparities seen in Alamos a and Grand Junction are attributed to large gaps between low income student performance for some of these students with district policies were difficult to overcome. The story is quite a bit different when looking at Pueblo East and North High School. Discussions with CGU staff focused on less than stellar performance from the motivator in importance of strong professionals in these roles

PAGE 95

81 Table IV.1 1 CGU Participating High School Graduation Rate and CGU Participant Graduation Rate College Aspiration Do students believe they will continue their education after high school? When both CGU participants and their peers were asked this question as seen in Figure IV.1, School # Cohort # Graduated Grad % Al l Students In Class All Student Grads All Student Grad Rate % CGU Higher Than School Grad Rate Abraham Lincoln 49 32 65.3% 471 299 63.5% 1.8% Alamosa 53 37 69.8% 148 128 86.5% 16.7% Aurora Central 74 35 47.3% 626 269 43.0% 4.3% Grand Junction Central 44 30 68.2% 434 341 78.6% 10.4% Greeley Central 17 14 82.4% 379 282 74.4% 8.0% Pueblo East 64 36 56.3% 223 163 73.1% 16.9% Lamar 58 38 65.5% 126 82 65.1% 0.4% Martin Luther King Jr 21 19 90.5% 82 75 91.5% 1.0% Montbello 23 15 65.2% 421 253 60.1% 5.1% North 60 28 46.7% 213 138 64.8% 18.1% Northridge 65 45 69.2% 295 216 73.2% 4.0% Wasson 41 21 51.2% 228 133 58.3% 7.1% Total 569 350 61.5% 3646 2379 65.2% 3.7%

PAGE 96

82 participants were more likely to state they would continue for a degree after high school. The first survey was taken after one year of services, durin g students 8 th grade year. The next year CGU student affirmative responses increased slightly, while control group students decreased slightly, but both percentages of affirmative answers were flat relative to the previous year. This can be attribu ted to students being three years, then two years away from high school graduation when students must have educational or career plans. Additionally, communication to students about college is more theoretical in younger high school grades, talking about w hat they could do with a postsecondary degree, as well as focused on their more brainstormed ideas about what they could accomplish with a degree. During their sophomore year, both participants and peers increased markedly, as this is when students begin to realize they must make objective and lasting decisions about their lives after high school. The junior year is spent sifting through collegiate or workforce options, and often the process of deciding on a path can deflate students, leaving them more re alistic, but also less excited about their collegiate plans. While both participants and peers increased their affirmative responses to whether they would continue their education, it was not as large an increase as seen during the sophomore year. Comparin g CGU participants and the control group, CGU students were more likely to state they would continue their education from the first administered survey through the final survey. They also saw a larger increase in affirmative responses than their peers.

PAGE 97

83 Figure IV. 3 CGU versus Control Group: Do You Think That You Will Go On For Further Education? For survey q uestion College Aspiration, you will go on to further education after high sch Table IV.1 2 Descriptive Statistics Research Question I, Survey Question College Aspiration GroupCode N Mean Std Deviation 8 th Grade Do you think that you will go on for further education after you leave high school? Control 90 4.20 .722 GEAR UP 1385 4.37 .775 11 th Grade Do you think you will go on for further education after high school? Control 1394 4.3 4 .825 GEAR UP 1182 4.59 .655

PAGE 98

84 Table IV.1 3 Independent Samples Test Research Question I, Survey Question College A spiration Levene's Test for Equality of Variances t test for Equality of Means 95% Confidence Interval of the Difference F Sig. t df Sig. (2 tailed) Mean Difference Std. Error Difference Lower Upper 8 th Grade Do you think that you will go on for further education after you leave high school? Equal variances assumed 3.662 .056 2.013 1473 .044 .169 .084 .334 .004 Equ al variances not assumed 2.141 102.774 .035 .169 .079 .325 .012 11 th Grade Do you think that you will go on for further education after you graduate from high school? Equal variances assumed 85.777 .000 8.383 2574 .000 .249 .030 .307 .191 E qual variances not assumed 8.541 2563.260 .000 .249 .029 .306 .192

PAGE 99

85 An independent samples t test was conducted to compare the attitudes of high school students participating in CGU and a control group of high school students regarding whether t hey think they will go on to further education after high school. This question was asked at two different time periods, first in the students 8 th grade year and then in the students 11 th grade academic year. There was no significant difference in the atti tude scores between CGU students (M=4.37, SD=.775) and the control group (M=4.2, SD=.722) in the 8 th grade year; t= 2.01 p=.0 44 In the second survey period there was a significant difference between the attitude scores of CGU students (M=4.59, SD=..655) a nd the control group (M=4.34, SD=.825) in the 11 th grade year; t= 8. 54 p=.000. College aspiration 8 th grade. Table IV. 1 4 Model Summary Research Question I, Survey Question College Aspiration 8 th Grade Model R R Square Adjusted R Square Std. Error of the Estimate 1 .166 a .028 .024 .382 Table IV.1 5 ANOVA Research Question I, Survey Question College Aspiration 8 th Grade Model Sum of Squares df Mean Square F Sig. 1 Regression 24.294 22 1.104 7.562 .000 a Residual 854.983 5855 .146 Total 879. 277 5877

PAGE 100

86 Table IV.1 6 Coefficients Research Question I, Survey Question College Aspiration 8 th Grade Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics B Std. Error Beta Tolerance VIF 1 (Constant) 4 .272 .029 148.289 .000 GroupCode .002 .017 .002 .091 .927 .343 2.917 gender .044 .014 .042 3.260 .001 .988 1.012 American Indian .008 .059 .002 .132 .895 .946 1.057 Asian .121 .047 .035 2.550 .011 .902 1.108 African American .069 .029 .035 2.335 .020 .747 1.339 Hispanic .012 .018 .015 .640 .522 .320 3.127 Native Hawaiian .048 .103 .006 .466 .642 .983 1.017 Not Reported .018 .068 .004 .271 .787 .968 1.033 White .005 .022 .004 .219 .827 .579 1.726 Freed Middle .171 .074 .032 2.301 .021 .842 1.187 Lamar HS .096 .383 .003 .250 .802 .994 1.006 Lamar Middle .197 .039 .093 5.094 .000 .499 2.005 Pueblo East HS .230 .193 .015 1.189 .234 .979 1.021

PAGE 101

87 AWCPA .088 .040 .040 2.211 .027 .496 2.014 Grand Mesa Middle .064 .045 .024 1.422 .155 .597 1.676 Franklin Middle .057 .042 .023 1.348 .178 .558 1.791 Ortega Middle .148 .042 .059 3.473 .001 .568 1.762 Colorado Springs East Middle .627 1.434 .202 .437 .662 .001 1287.548 Martin Luther King .132 .040 .381 3.252 .001 .012 82. 594 Kunsmiller Middle .114 .038 .329 2.952 .003 .013 74.643 Mt. Garfield Middle .029 .084 .085 .347 .729 .003 359.201 Jefferson Middle .037 .071 .108 .521 .602 .004 256.280

PAGE 102

88 A linear regression was performed to determine if the following in dependent or predictor variables: gender, ethnicity, middle school, and participant in Gear UP or not could be used to estimate a student's attitude regarding matriculation to postsecondary education. Using these predictor variables a significant model eme rged (F=7.56 p<.000 ). Despite the significance of the predictor model the Adjusted R square = .024, suggests low predictability strength and the variables in the model only account for under 5% of the variance in the survey question answer. Several indepe ndent variables in the model were significant including: gender Beta =.0 42 p=.0 01 A sian Beta = .0 35 p=.0 11 African American Beta = .035 p=.0 20 Freed Middle School Beta = .032 p=.0 21 Lamar Middle School Beta = .0 92 p=.0 00, AWCPA Beta = .027, p= 040, Ortega Middle School Beta = .059, p= .001, Martin Luther King Middle School Beta = .381, p=.001, and Kunsmiller Middle School Beta = .329, p=.003. These significant independent variables are contributing to a portion of the variance in the regression model. Holding all the other independent variables constant in the model, females compared to males tend to answer about .04 points more positively on the Likert scale regarding the likelihood they will go to college Asian students and African American s tudents compared to Hispanic students on average are likely to answer .04 more positive on the Likert scale regarding college aspiration. Again, 8 th grade females answered more positively than the 8 th grade males. Two middle schools results are particular ly interesting. The students at Martin Luther King Middle School answered in a significantly more positive way (.381) than their peers. It was during this time when the middle school was being transformed into a 6 12 early college. Just as striking is the numbers from Kunsmiller Middle School. Kunsmiller

PAGE 103

89 Middle School is the feeder middle school to Abraham Lincoln High School. During this time Lincoln was receiving a lot of attention for sending traditionally underserved students to college. It does not app ear that message had filtered down to the middle school as those students answered this question at a significantly lower rate ( .329). Collinearity diagnostics were run to determine if there were high correlations amongst the predictor variables. A tolera nce level of .01 was used. Applying these criteria three middle school falls below the .01 tolerance level therefore suggests there might be multicollinearity issue with the middle school variable. The collinearity between some of the middle schools makes it challenging to draw inferences about the relative contribution of each predictor variable to the overall success of the model. However, the middle school variable was not excluded from the model because it has practical implications. Again, the three m iddle schools that show to have a multicollinearity issue are the three schools associated with Colorado Springs; East Middle School, Jefferson Middle School and Garfield Middle School.

PAGE 104

90 College aspiration 11 th grade. Table IV.1 7 Model Summary Researc h Question I, Survey Question College Aspiration 11 th Grade Model R R Square Adjusted R Square Std. Error of the Estimate 1 .190 a .036 .033 .496 Table IV.1 8 ANOVA Research Question I, Survey Question College Aspiration 11 th Grade Model Sum of Squa res df Mean Square F Sig. 1 Regression 53.991 21 2.571 10.456 .000 a Residual 1439.944 5856 .246 Total 1493.934 5877

PAGE 105

91 Table IV.1 9 Coefficients Research Question I, Survey Question College Aspiration 11 th Grade Model Unstandardized Coeffici ents Standardized Coefficients t Sig. Collinearity Statistics B Std. Error Beta Tolerance VIF 1 (Constant) 4.369 .026 167.815 .000 GroupCode .205 .022 .201 9.156 .000 .341 2.935 gender .095 .018 .069 5.358 .000 .987 1.013 American Indian .06 5 .076 .011 .848 .397 .945 1.058 Asian .011 .061 .002 .183 .854 .906 1.104 African American .001 .038 .000 .029 .977 .753 1.329 Hispanic .144 .024 .138 6.058 .000 .317 3.150 Native Hawaiian .244 .134 .024 1.821 .069 .980 1.020 Not Reported .183 .088 .027 2.088 .037 .976 1.024 White .070 .028 .041 2.511 .012 .604 1.656 Lamar HS .003 .032 .002 .089 .929 .496 2.014 AWCPA .105 .114 .012 .923 .356 .955 1.047 Martin Luther King .105 .039 .043 2.712 .007 .651 1.536

PAGE 106

92 Aurora Centra l High .065 .033 .034 1.964 .050 .537 1.863 Grand Junction Central HS .059 .034 .030 1.743 .081 .541 1.848 Northridge HS .049 .033 .025 1.455 .146 .550 1.817 Greeley Central HS .040 .048 .012 .836 .403 .778 1.285 Alamosa HS .065 .034 .033 1.925 .054 .554 1.804 Wasson HS .019 .036 .009 .523 .601 .592 1.689 North HS .082 .035 .039 2.339 .019 .593 1.686 Montbello HS .004 .038 .002 .107 .915 .630 1.588 Abraham Lincoln HS .039 .033 .020 1.173 .241 .552 1.810

PAGE 107

93 A linear regression w as performed to determine if the following independent or predictor variables: gender, ethnicity, high school, and participant in Gear UP or not could be used to estimate a student's attitude regarding graduating from high school. Using these predictor var iables a significant model emerged (F=10.46, p<.000). Despite the significance of the predictor model the Adjusted R square = .033, suggests low predictability strength and the variables in the model only account for under 5% of the variance in the survey question answer. Several independent variables in the model were significant including: group code Beta =.201, p=.000, gender Beta = .069, p=.000, Hispanic Beta = .138, p=.000, White Beta = .041, p=.012, Martin Luther King Beta = .043, p=.007, Aurora Cen tral Beta = .034, p= 050 and North High School Beta = .039, p=.019. These significant independent variables are contributing to a portion of the variance in the regression model. Holding all the other independent variables constant in the model, a typica l CGU student is likely to answer .201 points more positively on the five point Likert scale than a student not in the CGU program regarding the likelihood they will attend college. This means if the control group mean score for this question is 4.34 the CGU student mean score would be 4.535 Holding all the other independent variables constant in the model, females compared to males tend to answer about .07 points more positively on the Likert scale regarding the likelihood they will attend college. This regression shows other variables being equal and controlled for that students in the 11 th grade believed much more strongly that they were going to attend college. This is strong evidence that the PCAs were very strong and that the combination of the knowl edge gleaned from the college pathway curriculum and the motivation from the PCAs significantly influenced student beliefs in their ability to go to college.

PAGE 108

94 Patterns are beginning to evolve. Again, there are significant differences between students who we re part of the CGU program answering more positively in the 11 th grade survey than their peers. Females are answering all questions more positively in each survey than the males. As for the ethnicity and school variables, it is difficult to draw strong con clusions. The two schools that bear mentioning are Martin Luther King and North High School. The students at Martin Luther King still answered the survey more positively than their peers, although the difference was not nearly as great. North High School w as in the throes of redesign discussions due to overall poor performance. The fact that students there answered this question at a lower rate than their peers might be indicative of what was occurring there. Collinearity diagnostics were run to determine i f there were high correlations amongst the predictor variables. A tolerance level of .01 was used. Applying these criteria no variables below the .01 tolerance level therefore suggests is not a multicollinearity issue with any of the variables. CSAP rat es and college enrollment. For students to develop aspirations to attend college, they must spend time goal setting and planning, and it takes self motivation to achieve these goals regardless of potential barriers. CGU students not only face socio econom ic barriers to attending postsecondary education, they also face academic barriers that make progression into college more difficult. Table IV.12 shows that 75.2% of CGU students who continued into their second semester of college scored at an unsatisfacto ry or partially proficient level in at least one subject on the Colorado State Assessment Program during 8 th grade.

PAGE 109

95 These students were not meeting grade level expectations and had to improve toward a proficient level in English math, reading, or science by the time they graduated high school. In addition to working through class requirements in all their high school classes, they had to complete additional work to be prepared to succeed in college, proving they were willing to work through barriers to mee CGU credit for good work that was clearly done by schools and teachers. This is just a point of reference and should be utilized as a place in which further study should occur. Table IV. 1 9 depicts the breakdow n of CGU college enrollment. Of the 214 students who matriculated to a postsecondary institution, 75% of those students the program. Ten of those students went on to selective four year institutions. This would appear to be significant as these students were considered behind going into their 8 th grade year of school. This could indicate that CGU staff provided proper mot ivation and that the schools in which these students attended did a great job of helping these students catch up. At bare minimum, it shows that CGU worked with students who were not necessarily seen as college material at that point in their academic care ers.

PAGE 110

96 Table IV. 20 CGU Participant College Enrollment Type and Colorado Student Assessment Program Completion Levels College Type Population At Least One U % At Least One U At Least One PP % At Least One PP Either % Any Deficiency Selective Four Year 10 2 20.0% 5 50.0% 5 50.0% Four Year 96 18 18.8% 59 61.5% 63 65.6% Two Year 108 48 44.4% 88 81.5% 93 86.1% Grand Total 214 68 31.8% 152 71.0% 161 75.2%

PAGE 111

97 CGU vs. National Data Students from 12 high schools were surveyed by the CGU program during the 11 th grade Of these students, 97.3% graduate from high school. The 2009 High School Survey of Student Engagement surveyed ( N=42,754) students and had 91.4% expected to at tain at least a high school diploma (Yazzie Mintz, 2010). Students were also asked whether they would attain some kind of postsecondary education and 87% responded they would like to. Of the students within the CGU program that w ere surveyed in the 11th grade 93.9% seek out further education upon completion of high school. The percentages are noticeably higher for the students within the CGU program. This is even more significant considering four of the Colorado schools with CGU programs (Abraham Lincoln, MLK, Montbello, and North) are part of the Denver Public School s ystem, which in 2010 had a 51.8% high school graduation rate. The Colorado gradua tion rate is approximately 72.4% (Robles, 2010 Denv er Post) and the national gr aduation rate is 68.8% (Alliance for Excellent Education, 2009). The results from the 11 th grade survey can also be compared to the 8 th grade survey of middle school students. Eight grade students from 14 middle schools answer ed e students that responded, 94.3% stated they would definitely or probably graduate from high school. This number has not been realized when the national and state graduation rates are considered. Of the CGU participants that were survey ed in 11 th grade 97.3% stated that they would definitely or probably graduate from high school. Participation in CGU

PAGE 112

98 improves the chances that the goals and perceptions of students at the middle s chool level are met once they reach high school. Do you think that you will go on for further education after you leave high school? Of those students that responded to ans yed in middle school only 64.86% of students could answer that they would definitely or probably continue on fo r additional education upon the completion of high sch ool. When compared to the 93.9% of the CGU participants that answered that they would definitely or probably attain additional education it may be concluded that their participation improved their cha nces of graduation and motivation to continue on to higher education. Research Question II: CGU PCA Impact on Student Interest and Conviction in Postsecondary Education. CGU vs. National Data Evidence from research notes the importance of relationships within schools. Yazzie Mintz (2010) notes that a connection to at least one adult is significantly environment. The 2009 High School Survey of Student Engagement found that 82% of students agreed or strongly agreed that they were supported by teachers, 65% by administrators, and 74% by counselors (Yazzie Mintz, 2010). Of the total number of students responding to survey question on High School Class Requirements in the 11 th grade survey, 89.7% stated that a mentor at their high school had an influence on them.

PAGE 113

99 While survey questions utilized in Research Question II did ask specifically about elors related to information they received from CGU PCAs or other mentors. This would be an excellent area for further research to see how the layering of information f rom different school advocates, as well as the relationships of these advocates with one another, could impact their aspiration for postsecondary education. PCA Relationship College Ability The s urvey question that addressed the relationship of the PCA to a belief in their ability to attend college was chosen to help understand the impact that a PCA from the CGU program had on their individual students. This relationship was viewed in a couple of different ways. Because it was very difficult to address the relationship between the control group and their counselors in any meaningful way in the 8 th grade (there is an extremely high ratio of students to counselors: in most of the CGU partner schools there was only one counselor for the entire scho at all) a statistical comparison was only used for the 11 th grade cohort of CGU students. The survey question pertaining to College Ability asked for you? Rate how much you agree or disagree with the following statement : It will help me go to college. Table IV.2 1 Descriptive Statistics Research Question II, Survey Que stion PCA Relationship College Ability GroupCode N Mean Std.

PAGE 114

100 Deviation 8 th Grade What is your GEAR UP PCA doing for you? Rate how much you agree or disagree with the following statement: She/he will help me be able to go to college. Control 88 3.1 4 .761 GEAR UP 1380 3.67 .558 11 th Grade What is your GEAR UP PCA doing for you? Rate how much you agree or disagree with the following statement: She/he will help me be able to go to college. Control 1372 3.12 .678 GEAR UP 1172 3.48 .558

PAGE 115

101 Table IV.2 2 Independent Samples Test Research Question II, Survey Question PCA Relationship College Ability Levene's Test for Equality of Variances t test for Equality of Means 95% Confidence Interval of the Difference F Sig. t df Sig. (2 ta iled) Mean Difference Std. Error Difference Lower Upper 8 th Grade What is the GEAR UP program doing for you ? It will help me be able to go to college. Equal variances assumed 5.286 .022 8.449 1466 .000 .532 .063 .655 .408 Equal variances not assum ed 6.446 93.072 .000 .532 .082 .696 .368 11 th Grade What is your school doing for you ? It will help me be able to go to college Equal variances assumed 6.662 .010 14.556 2542 .000 .362 .025 .411 .313 Equal variances not assumed 14.778 2 538.428 .000 .362 .025 .410 .314

PAGE 116

102 An independent samples t test was conducted to compare how much students get into college and a control group of high school student s and their relationship with their traditional school counselor would help them get into college. This question was asked at two different time periods, first in the students 8 th grade academic year and then in the students 11 th grade academic year. There was significant difference in the attitude scores between CGU students (M=3.67 SD=.558) and the control group (M=3.14, SD=.761) in the 8 th grade year; t=. 6.446 p=.000 In the second survey period there was a significant difference between the attitude sc ores of CGU students (M=3.48, SD=.558) and the control group (M=3.12, SD=.678) on the 11 th grade survey ; t= 14.78, p=.00 0. PCA relationship college ability 8 th grade. Table IV.2 3 Model Summary Research Question II, Survey Question PCA Relationship College Ability 8 th Grade Model R R Square Adjusted R Square Std. Error of the Estimate 1 .252 a .063 .060 .284 Table IV.2 4 ANOVA Research Question II, Survey Question PCA Relationship College Ability 8 th Grade Model Sum of Squares df Mean Square F Sig. 1 Regression 31.873 22 1.449 17.976 .000 a Residual 471.879 5855 .081 Total 503.752 5877

PAGE 117

103 Table IV.2 5 Coefficients Research Question II, Survey Question PCA Relationship College Ability 8 th Grade Model Unstandardized Coefficient s Standardized Coefficients t Sig. Collinearity Statistics B Std. Error Beta Tolerance VIF 1 (Constant) 3.709 .021 173.268 .000 GroupCode .003 .013 .005 .218 .827 .343 2.917 gender .023 .010 .028 2.239 .025 .988 1.012 American Indian .015 .044 .004 .344 .731 .946 1.057 Asian .000 .035 .000 .002 .999 .902 1.108 African American .003 .022 .002 .122 .903 .747 1.339 Hispanic .005 .014 .008 .365 .715 .320 3.127 Native Hawaiian .073 .077 .012 .952 .341 .983 1.017 Not Reported 060 .050 .015 1.188 .235 .968 1.033 White .004 .016 .004 .218 .827 .579 1.726 Freed Middle .533 .055 .133 9.663 .000 .842 1.187 Lamar HS .074 .285 .003 .261 .794 .994 1.006 Lamar Middle .044 .029 .027 1.524 .128 .499 2.005 Pueblo East H S .041 .144 .004 .283 .777 .979 1.021

PAGE 118

104 AWCPA .027 .030 .016 .913 .361 .496 2.014 Grand Mesa Middle .228 .034 .111 6.775 .000 .597 1.676 Franklin Middle .107 .031 .058 3.414 .001 .558 1.791 Ortega Middle .016 .032 .008 .500 .617 .568 1.76 2 Colorado Springs East Middle 12.204 1.065 5.200 11.458 .000 .001 1287.548 Martin Luther King .057 .030 .219 1.902 .057 .012 82.594 Kunsmiller Middle .115 .029 .439 4.021 .000 .013 74.643 Mt. Garfield Middle .740 .063 2.838 11.838 .000 .0 03 359.201 Jefferson Middle .452 .053 1.734 8.563 .000 .004 256.280

PAGE 119

105 A linear regression was performed to determine if the following independent or predictor variables: gender, ethnicity, middle school, and participant in Gear UP or not could be us e d to PCA. Using these predictor variables a significant model emerged (F= 17.98, p = .000 ). Despite the significance of the predictor model the Adjusted R square = .0 60 suggests low predi ctability strength and the variables in the model only account for under 5% of the variance in the survey question answer. Several independent variables in the model were significant including: gender Beta =.0 25 p=.0 28 Freed Middle School Beta = .133 p= .0 00 Grand Mesa Middle School Beta = .111, p=. 000 Franklin Middle School Beta = 058 p=. 001 and Kunsmiller Middle School Beta = .439, p= .000. These significant independent variables are contributing to a portion of the variance in the regression mo del. Holding the other entire independent variables constant in the model, females compared to males tend to answer about .03 points more positively on the Likert scale regarding the role that they feel their PCA will play in helping them attend college. Grand Mesa, Franklin, and Kunsmiller middle schools all answered significantly lower than their peers. Collinearity diagnostics were run to determine if there were high correlations amongst the predictor variables. A tolerance level of .01 was used. Apply ing this criterion the three schools associated with Colorado Springs fall below the .01 tolerance level therefore suggests there might be multicollinearity issue with the middle school variable, or a fidelity issue with the delivery in those schools.

PAGE 120

106 PCA relationship college ability 11 th grade. Table IV.2 6 Model Summary Research Question II, Survey Question PCA Relationship College Ability 11 th Grade Model R R Square Adjusted R Square Std. Error of the Estimate 1 .239 a .057 .054 .417 Tabl e IV.2 7 ANOVA Research Question II, Survey Question PCA Relationship College Ability 11 th Grade Model Sum of Squares df Mean Square F Sig. 1 Regression 61.844 21 2.945 16.965 .000 a Residual 1016.542 5856 .174 Total 1078.386 5877

PAGE 121

107 Table IV.2 8 Coefficients Research Question II, Survey Question PCA Relationship College Ability 11 th Grade Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics B Std. Error Beta Tolerance VIF 1 (Constant) 3.213 .0 22 147.901 .000 GroupCode .261 .019 .302 13.881 .000 .341 2.935 gender .035 .015 .030 2.347 .019 .987 1.013 American Indian .168 .064 .034 2.622 .009 .945 1.058 Asian .214 .052 .055 4.158 .000 .906 1.104 African American .140 .032 .064 4.393 .000 .753 1.329 Hispanic .132 .020 .149 6.601 .000 .317 3.150 Native Hawaiian .083 .113 .009 .740 .460 .980 1.020 Not Reported .133 .074 .023 1.811 .070 .976 1.024 White .130 .024 .090 5.530 .000 .604 1.656 Lamar HS .015 .027 .01 0 .541 .589 .496 2.014 AWCPA .135 .096 .018 1.409 .159 .955 1.047 Martin Luther King .163 .033 .078 4.984 .000 .651 1.536 Aurora Central High .060 .028 .038 2.175 .030 .537 1.863

PAGE 122

108 Grand Junction Central HS .033 .029 .020 1.161 .246 .541 1.848 Northridge HS .043 .028 .026 1.526 .127 .550 1.817 Greeley Central HS .042 .041 .015 1.028 .304 .778 1.285 Alamosa HS .043 .028 .026 1.533 .125 .554 1.804 Wasson HS .024 .030 .013 .812 .417 .592 1.689 North HS .005 .030 .003 .177 .860 .593 1.6 86 Montbello HS .063 .032 .032 2.010 .044 .630 1.588 Abraham Lincoln HS .044 .028 .027 1.567 .117 .552 1.810

PAGE 123

109 A linear regression was performed to determine if the following independent or predictor variables: gender, ethnicity, high school, and p articipant in Gear UP or not could be use d to estimate a student's attitude regarding their relationship with their PCA and that relationship to help matriculation to college Using these predictor variables a significant model emerged (F= 16.97 p<.000). Despite the significance of the predictor model the Adjusted R square = .0 54 suggests low predictability strength and the variables in the model only account for under 5% of the variance in the survey question answer. Several independent variables in the model were significant including: group code Beta =. 302 p=.0 00 gender Beta = .0 30 p=.0 19 American Indian Beta = .034, p=.009 Asian Beta = 055, p=.000 Hispanic Beta = .149 p=.0 00, White Beta = .090, p= .000, Martin Luther King Beta = .078, p=.00 0, Aurora Central Beta = .038, p= .032, and Montbello Beta = .032, p= .044. These significant independent variables are contributing to a portion of the variance in the regression model. Holding all the other independent variables constant in the model, a typical CGU student is likely to answer .302 points more positively on the five point Likert scale than a student not in the CGU program regarding whether their PCA will help them attend college. This means if the control group mean score for this quest ion is 3. 12, the CGU student mean would be 3.422 Holding all the other independent variables constant in the model, females compared to males tend to answer .03 points more positively on the Likert scale regarding the likelihood their PCA will help them t o attend college. This regression proves that all other variables being equal and controlled for those students in the CGU program believed much more positively that their PCA would help them get to college as

PAGE 124

110 it related to their peers beliefs that their s chool counselor would help them. This highlights the huge student to counselor ratio in high schools. The significantly higher positive answer from CGU students on this question is telling. It indicates that a strong relationship with a PCA is very importa nt in helping students believe that they can attend college. The other two variables of note are the lower scores from Montbello and Aurora Central. In discussions with CGU administration, both PCAs at those schools were strong and had committed. Collinear ity diagnostics were run to determine if there were high correlations amongst the predictor variables. A tolerance level of .01 was used. Applying these criteria no variables fall below the .01 tolerance level therefore suggests there is not a multicollin earity issue with any of the variables

PAGE 125

111 GEAR UP retention rates and class performance. How do CGU students perform once they attend a postsecondary institution? Does building up their aspirations to attend college change their resiliency once they are on semester? The answers to these questions define much of the positive impact CGU services have on students. To aid student transitions into college, CGU provided staff to area colleges and universities to follow CGU students and provide guidance and support for them. In Table IV. 28 the college retention rate from first semester to second semester students that are both continuing at their current college and those that are continuing with their post high school plan, two percentages are listed. The first shows the CGU retention rate at their current college, and the second shows students that is e ither continuing at their current college, transferring to a new college, or joining the military. Of the 39 colleges CGU students attend, only three have higher retention rates for their freshman class than the CGU retention rate. Not including transfer s or military service, seven of 29 schools have higher class retention rates than the CGU cohort. Not only did 78% of active CGU students enroll in postsecondary institutions, but they persisted at their chosen institution at higher rates than their classm ates, offering evidence that CGU services provided by the institution. There are a couple of ways to look at this. As is true with all of this data, CGU has a fair ly small sample size and gleaning trends or measuring success should be understood with that backdrop. Certainly in some of the schools listed, there are only a few students.

PAGE 126

112 Still, especially in the community colleges, where students who attend often have less background and support, the numbers shine favorable on CGU. In discussions with CGU students knew more about what to do upon matriculation, had more confidence due to success in dual statements about the effects that CGU had with college retention, but the initial data shows a promising trend. Table IV. 2 9 CGU Participant and Full College Population Retention Rates College College Retention Rate GU Retention Rate Including Transfers & Military % GU Higher Than School Retention Rate Adams State College 80% 72% 91% 11% Aims Co mmunity College 52% 76% 82% 30% Colorado Mesa University 67% 52% n/a 15% Colorado Mountain College 81% 75% 100% 19% Colorado Northwestern 67% 100% n/a 33% Colorado School Of Mines 87% 50% 100% 13% Colorado State University 83% 93% 100% 17% Colorado State University Pueblo 64% 86% 93% 29% Community College Of Aurora 50% 72% 81% 31% Community College Of Denver 55% 70% 80% 25%

PAGE 127

113 Emily Griffith 60% 67% n/a 7% Fort Hays State University 100% 100% 100% Fort Lewis Co llege 62% 75% n/a 13% Front Range Community College 57% 100% n/a 43% Hamline University 100% 100% 100% Hastings College 73% 100% 100% 27% Imperial Valley 74% 0 100% 26% Lamar Community College 52% 75% n/a 23% Metrop olitan State Clg Of Denver 66% 78% 86% 20% New Mexico Highlands 0 100% 100% New York University 92% 100% 100% 8% Northeastern Junior College 55% 83% 100% 45% Otero Junior College 100% 100% 100% Pickens Tech Ctr Aurora Pub Sch 69% 50% n/ a 19% Pikes Peak Community College 54% 62% 75% 21% Pueblo Community College 56% 76% n/a 20% Red Rocks Community College 54% 100% 100% 46% Redstone College 66% 100% 100% 34% Regis University 93% 100% 100% 7% Texas Tech University 80% 100% 100% 20% Trinidad State Jr College 62% 87% n/a 25% University Of Arizona 77% 100% 100% 23% University Of Colo / Boulder 84% 100% 100% 16% University Of Colo /Colo Springs 68% 71% 86% 18% University Of Colorado Denver 73% 89% 100% 27% Universit y Of Denver 87% 100% 100% 13% University Of Northern Colorado 69% 67% 75% 6% University of Wyoming 72% 100% 100% 28%

PAGE 128

114 Utah State University 72% 33% 67% 5% Students have the motivation and resiliency to persist at their chosen school. How are they doi ng in their classes? This provides a less definitive picture as to their current success and likelihood to continue, as viewed in Table IV. 29 CGU students are passing their classes at a higher rate than their classmates at nine of the 24 schools where the se statistics could be gathered. Additionally, they are passing their classes at a higher rate than their classmates with at least a C or better at nine of the 24 schools where these statistics could be gathered. Students are broadly not performing as well in their classes as their classmates at all but nine schools. However, these are students that have already exhibited a willingness to work through barriers to enrollment, and it is possible that the services provided by CGU will push them through potenti al issues arising from lower grades such as probation, retaking classes, or an inability to select or continue in their chosen majors. It should be noted that GEAR UP is not a program that aspires to impact students academic ability, which is left to th e schools. Rather, it works to motivate students to see postsecondary education as their goal and to prepare them to succeed once they get there. While this delineation is gray and has many aspects of the program delving into academic preparation and conte nt knowledge including remedial courses and concurrent enrollment, CGU advisors do not teach courses or tutor in academic areas. The program provides the funding for students to take advantage of these opportunities rather than administering the courses an d providing the content. Two things to be noted here; again, a small sample size overall and especially in a couple of these schools. That small sample size tends to skew the look of the data

PAGE 129

115 (positively or negatively) in a pretty dramatic way. Second th ing to note, and maybe a lot more important, is the pass rates that area colleges are reporting. They are dramatically higher for first time freshman (with whom it might be expected would struggle the most) than their 4 year graduation rates. This disparit y could be the basis for future study.

PAGE 130

116 Table IV. 30 CGU Participant and Full College Population Course Performance Rates College GU Pass Rate College Overall Pass Rate % GU Higher Than School Pass Rate GU C or Higher Pass Rate College C or Higher Pass Rat e % GU Higher Than School C or Higher Pass Rate Adams State College 56% 83% 27% 35% 65% 30% Aims Community College 79% 75% 4% 65% 62% 3% Colorado Mesa University 48% 82% 34% 41% 65% 24% Colorado Mountain College 78% 81% 3% 67% 68% 1% Colorado No rthwestern Community College 100% 62% 38% 100% 55% 45% Colorado School Of Mines 100% 98% 2% 92% 89% 3% Colorado State University 91% 92% 1% 75% 81% 6% Colorado State University Pueblo 76% 84% 8% 59% 66% 7% Community College Of Aurora 58% 54% 4% 52% 45% 7% Community College Of Denver 41% 44% 3% 35% 35% 0% Emily Griffith Technical College 81% N/A N/A 81% N/A N/A Fort Hays State University 100% N/A N/A 100% N/A N/A

PAGE 131

117 Fort Lewis College 62% 91% 29% 37% 78% 41% Front Range Community College 20% 69% 49% 0% 57% 57% Hamline University 100% N/A N/A 80% N/A N/A Hastings College 71% N/A N/A 71% N/A N/A Imperial Valley College 100% N/A N/A 100% N/A N/A Intellitec College 0% N/A N/A 0% N/A N/A Lamar Community College 44% 85% 41% 38% 70% 32% Metrop olitan State Clg Of Denver 53% 83% 30% 45% 70% 25% New Mexico Highlands University 0% N/A N/A 0% N/A N/A New York University 86% N/A N/A 0% N/A N/A Northeastern Junior College 80% 70% 10% 78% 59% 19% Otero Junior College 100% 65% 35% 100% 56% 44% Pi ckens Tech Ctr Aurora Pub Sch 100% N/A N/A 100% N/A N/A Pikes Peak Community College 30% 62% 32% 0% 51% 51% Pueblo Community College 60% 62% 2% 53% 50% 3% Red Rocks Community College 70% 64% 6% 50% 54% 4% Restone 50% N/A N/A 50% N/A N/A Regis Univ ersity 100% N/A N/A 100% N/A N/A

PAGE 132

118 Texas Tech University 100% N/A N/A 100% N/A N/A Trinidad State Jr College 84% 84% 0% 81% 71% 10% University Of Arizona 100% N/A N/A 100% N/A N/A University Of Colorado At Boulder 100% 95% 5% 78% 84% 6% University Of C olorado At Colo Springs 73% 93% 20% 0% 82% 82% University Of Colorado Denver 90% 91% 1% 83% 80% 3% University Of Northern Colorado 45% 88% 43% 38% 72% 34% University Of Wyoming 100% N/A N/A 100% N/A N/A Utah State University 93% N/A N/A 93% N/A N/ A Overall 63% N/A N/A 54% N/A N/A 4 Year 68% N/A N/A 56% N/A N/A 2 Year 60% N/A N/A 54% N/A N/A Out of State 85% n/a n/a 80% n/a n/a

PAGE 133

119 Research Question III: CGU Impact on Low Income Student Knowledge and Understanding of Postsecondary Pathways CGU vs. Control CGU participants were asked what the program was doing for them in relation to helping them with accessing higher education. The control group was asked if their school was providing assistance in helping them access higher education in survey q uestion s on courses needed to best prepare a student for college and college entrance requirements. These are similar but not the same. The first is specific to the pathway courses that students take in high school and their rigor. The second is specific t o the expectations colleges have upon student matriculation. Unfortunately the two do not always match up. High School Class Requirements Table IV. 3 1 Descriptive Statistics of Research Question III Survey Question High School Class Requirements GroupCode N Mean Std. Deviation 8 th Grade Has anyone from your school or GEAR UP ever spoken to you about Classes you need to take in high school to prepare for college? Control 90 2.18 .572 GEAR UP 1389 2.01 .557 11 th Grade Has anyone from your school ever spoken to you about the following: Classes you need to take in high school to prepare for college Control 1388 1.90 .503 GEAR UP 1182 2.02 .269

PAGE 134

120 Table IV.3 2 Independent Samples Test of Research Question III Survey Question High School Class Requirements Levene's Test for Equality of Variances t test for Equality of Means 95% Confidence Interval of the Difference F Sig. t df Sig. (2 tailed) Mean Differenc e Std. Error Difference Lower Upper 8 th Grade Has anyone from your school or GEAR UP ever spoken to you about Classes you need to take in high school to prepare for college? Equal variances assumed 6.390 .012 2.837 1477 .005 .172 .061 .053 .291 Equal variances not assumed 2.768 100.216 .007 .172 .062 .049 .295 11 th Grade Has anyone from your school ever spoken to you about the following: Classes you need to take in high school to prepare for college Equal variances assumed 309.833 .000 7.173 2568 .000 .117 .016 .149 .085 Equal variances not assumed 7.500 2184.48 2 .000 .117 .016 .148 .086

PAGE 135

121 An independent samples t test was conducted to compare the knowledge of courses needed to take in high school to be prepared for college. This ques tion was asked at two different time periods, first in the 8 th grade academic year and then in the 11 th grade academic year. There was a significant difference in the scores between CGU students (M=2.01, SD=.557) and the control group (M=2.18, SD= 572) in the 8 th grade year; t=2. 77 p= 0 07 In the second survey period there was a significant difference between the attitude scores of Gear Up students (M=2.02, SD=.269) and the control group (M=1.9, SD=.503) in the 11 th grade academic year; t= -7. 50 p=.000. H igh school class requirements 8 th grade. Table IV.3 3 Model Summary Research Question III Survey Question High School Class Requirements 8 th Grade Model R R Square Adjusted R Square Std. Error of the Estimate 1 .125 a .016 .012 .279 Table IV.3 4 AN OVA Research Question III Survey Question High School Class Requirements 8 th Grade Model Sum of Squares df Mean Square F Sig. 1 Regression 7.161 22 .325 4.194 .000 a Residual 454.450 5855 .078 Total 461.611 5877

PAGE 136

122 Table IV.3 5 Coefficients Research Question III Survey Question High School Class Requirements 8 th Grade Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics B Std. Error Beta Tolerance VIF 1 (Constant) 3.213 .022 147.901 .000 GroupCode .261 .019 .302 13.881 .000 .341 2.935 gender .035 .015 .030 2.347 .019 .987 1.013 American Indian .168 .064 .034 2.622 .009 .945 1.058 Asian .214 .052 .055 4.158 .000 .906 1.104 African American .140 .032 .064 4.393 .000 .753 1.3 29 Hispanic .132 .020 .149 6.601 .000 .317 3.150 Native Hawaiian .083 .113 .009 .740 .460 .980 1.020 Not Reported .133 .074 .023 1.811 .070 .976 1.024 White .130 .024 .090 5.530 .000 .604 1.656 Lamar HS .015 .027 .010 .541 .589 .496 2.01 4 AWCPA .135 .096 .018 1.409 .159 .955 1.047 Martin Luther King .163 .033 .078 4.984 .000 .651 1.536 Aurora Central High .060 .028 .038 2.175 .030 .537 1.863

PAGE 137

123 Grand Junction Central HS .033 .029 .020 1.161 .246 .541 1.848 Northridge HS .043 .028 .026 1.526 .127 .550 1.817 Greeley Central HS .042 .041 .015 1.028 .304 .778 1.285 Alamosa HS .043 .028 .026 1.533 .125 .554 1.804 Wasson HS .024 .030 .013 .812 .417 .592 1.689 North HS .005 .030 .003 .177 .860 .593 1.686 Montbello HS .0 63 .032 .032 2.010 .044 .630 1.588 Abraham Lincoln HS .044 .028 .027 1.567 .117 .552 1.810

PAGE 138

124 A linear regression was performed to determine if the following independent or predictor variables: gender, ethnicity, middle school, and participant in Gear U P to understand if they had understanding of courses needed to be eligible for college. Using these predictor variables a significant model emerged (F=4.19, p =. 000). Despite the significance of the predictor model the Adjusted R square = .012, suggests low predictability strength and the variables in the model only account for under 5% of the variance in the survey question answer. Several independent variables in the model were significant including: White Beta =. 036, p=.033, Lamar Beta = .053, p=.004, F ranklin Beta = 039, p=.02 6, Ortega Middle School Beta = 041 p=. 016 and Kunsmiller Mid dle School Beta = .305, p=.006 These significant independent variables are contributing to a portion of the variance in the regression model. Acknowledging the fac t that the scores from the schools associated with Colorado Springs are not to be relied on the answers from students at Franklin Middle School and Kunsmiller Middle School again show lower expectations and aspirations than their peers. Of the other varia bles that show significance, there is little in the way of generalizability. Collinearity diagnostics were run to determine if there were high correlations amongst the predictor variables. A tolerance level of .01 was used. Other than the three schools in Colorado Springs there is not a concern with multi collinearity.

PAGE 139

125 High school class requirements 11 th grade. Table IV.3 6 Model Summary Research Question III Survey Question High School Class Requirements 11 th Grade Model R R Square Adjusted R Squar e Std. Error of the Estimate 1 .129 a .017 .013 .274 Table IV.3 7 ANOVA Research Question III Survey Question High School Class Requirements 11 th Grade Model Sum of Squares df Mean Square F Sig. 1 Regression 7.427 21 .354 4.726 .000 a Residual 438 .201 5856 .075 Total 445.628 5877

PAGE 140

126 Table IV.3 8 Coefficients Research Question III Survey Question High School Class Requirements 11 th Grade Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics B Std Error Beta Tolerance VIF 1 (Constant) 1.915 .014 134.255 .000 GroupCode .087 .012 .157 7.076 .000 .341 2.935 gender .002 .010 .003 .239 .811 .987 1.013 American Indian .119 .042 .038 2.830 .005 .945 1.058 Asian .002 .034 .001 .070 .944 906 1.104 African American .077 .021 .055 3.669 .000 .753 1.329 Hispanic .042 .013 .073 3.179 .001 .317 3.150 Native Hawaiian .005 .074 .001 .066 .947 .980 1.020 Not Reported .090 .048 .025 1.870 .062 .976 1.024 White .045 .015 .048 2.889 .004 .604 1.656 Lamar HS .032 .018 .034 1.828 .068 .496 2.014 AWCPA .081 .063 .017 1.298 .194 .955 1.047 Martin Luther King .030 .021 .022 1.386 .166 .651 1.536

PAGE 141

127 Aurora Central High .026 .018 .025 1.410 .158 .537 1.863 Grand Junction Centr al HS .035 .019 .033 1.888 .059 .541 1.848 Northridge HS .009 .018 .008 .472 .637 .550 1.817 Greeley Central HS .043 .027 .024 1.606 .108 .778 1.285 Alamosa HS .015 .019 .014 .813 .416 .554 1.804 Wasson HS .012 .020 .010 .616 .538 .592 1.689 Nort h HS .041 .019 .035 2.095 .036 .593 1.686 Montbello HS .065 .021 .051 3.140 .002 .630 1.588 Abraham Lincoln HS .060 .018 .057 3.272 .001 .552 1.810

PAGE 142

128 A linear regression was performed in SPSS to determine if the following independent or predictor vari ables: gender, ethnicity, high school, and participant in Gear UP or not could be use d to determine if they had understood the courses needed to prepare them for graduation. Using these predictor variables a significant model emerged (F=4.73, p<.000). Des pite the significance of the predictor model the Adjusted R square = .013, suggests low predictability strength and the variables in the model only account for under 5% of the variance in the survey question answer. Several independent variables in the mod el were significant including: group Beta =. 157 p=.0 00 American Indian Beta = 038 p=.0 05 African American Beta = 055 p=. 000 Hispanic Beta = 073, p=.001 White Beta = .0 48 p=.0 04, North High School Beta = .035, p=.36, Montbello High School Beta = .051, p=. 002 and Lincoln High School Beta = .057, p=.001. These significant independent variables are contributing to a portion of the variance in the regression model. Holding all the other independent variables constant in the model, a typical CGU stu dent is likely to answer .157 points more positively on the five point Likert scale than a student not in the CGU program regarding the likelihood they understood high school graduation requirements. This means if the control group mean score for this ques tion is 1.90 the CGU student mean would be 2.057. This regression illuminates those students in the CGU program answered more positively than their peers as to knowing about graduation requirements. This concept is taught on several occasions each year as part of the GEAR UP curriculum and also tends to come up quite a bit in conversations between PCAs and their students when discussion dual enrollment opportunities.

PAGE 143

129 Other than the fact that students who were part of the CGU program again answered more pos itively to this question from the survey than their peers, there does not seem to be much in the way of generalizable information here. It is interesting to note however, that two schools (North and Montbello) had positive results on this question. Interes ting due to the fact that North had lower scores in several other questions, and that CGU administration had conc erns with the PCA at the school, Montbello for the fact that, by almost all other measures, the school climate was in disarray. CGU administrat ion did feel strongly that the PCA assigned to Montbello was especially strong. Collinearity diagnostics were run to determine if there were high correlations amongst the predictor variables. A tolerance level of .01 was used. Applying these criteria ther e is not one variable that falls below the .01 tolerance level therefore suggests there is not a multicollinearity issue with any of the variables. College Entrance Requirements Table IV. 3 9 Descriptive Statistics Research Question III S urvey Q uestion C ollege Entrance Requirements GroupCode N Mean Std. Deviation 8 th Grade Has anyone from your school or GEAR UP ever spoken to you about College entrance requirements? Control 90 2.04 .763 GEAR UP 1389 2.02 .608 11 th Grade H as anyone from your scho ol ever spoken to you about the following: College entrance requirements Control 1380 1.88 .638 GEAR UP 1176 2.07 .362

PAGE 144

130 Table IV. 40 Independent Samples Test Research Question III Survey Question College Entrance Requirements Levene's Test for Equ ality of Variances t test for Equality of Means 95% Confidence Interval of the Difference F Sig. t df Sig. (2 tailed) Mean Difference Std. Error Difference Lower Upper 8 th Grade Has anyone from your school or GEAR UP ever spoken to you about College entrance requirements? Equal variances assumed 17.451 .000 .436 1477 .663 .029 .067 .103 .161 Equal variances not assumed .357 96.456 .722 .029 .082 .134 .192 11 th Grade Has anyone from your school ever spoken to you about the following: College entrance requirements? Equal variances assumed 378.570 .000 9.067 2554 .000 .190 .021 .232 .149 Equal variances not assumed 9.447 2241.36 8 .000 .190 .020 .230 .151

PAGE 145

131 An independent samples t test was conducted to compare the knowledg e of courses needed to take in high school to be prepared for college. This question was asked at two different time periods, first in the 8 th grade academic year and then in the 11 th grade academic year. There was not a significant difference in the score s between CGU students (M=2.02, SD=.608) and the control group (M=2.04, SD=.763) in the 8 th grade year; t=. 357 p= .722 In the second survey period there was a significant difference between the scores of CGU students (M=2.07, SD=.362) and the control grou p (M=1.88, SD=.638) in the 11 th grade academic year; t= 9. 45 p=.000. College entrance requirements 8 th grade. Table IV. 4 1 Model Summary Research Question 3, Survey Question College Entrance Requirements 8 th Grade Model R R Square Adjusted R Square St d. Error of the Estimate 1 .095 a .009 .005 .309 Table IV. 4 2 ANOVA Research Question 3, Survey Question College Entrance Requirements 8 th Grade Model Sum of Squares df Mean Square F Sig. 1 Regression 5.118 22 .233 2.435 .000 a Residual 559.460 5855 .096 Total 564.577 5877

PAGE 146

132 Table IV.4 3 Coefficients Research Question III Survey Question College Entrance Requirements 8 th Grade Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics B Std. Error Beta Tolerance VIF 1 (Constant) 2.065 .023 88.606 .000 GroupCode .004 .014 .006 .280 .780 .343 2.917 gender .001 .011 .001 .083 .934 .988 1.012 American Indian .003 .048 .001 .070 .944 .946 1.057 Asian .041 .038 .015 1.072 .284 .902 1.10 8 African American .018 .024 .012 .766 .444 .747 1.339 Hispanic .011 .015 .017 .758 .448 .320 3.127 Native Hawaiian .004 .083 .001 .051 .960 .983 1.017 Not Reported .000 .055 .000 .005 .996 .968 1.033 White .021 .018 .020 1.192 .233 .579 1 .726 Freed Middle .001 .060 .000 .021 .983 .842 1.187 Lamar HS .055 .310 .002 .176 .860 .994 1.006 Lamar Middle .055 .031 .033 1.767 .077 .499 2.005 Pueblo East HS .194 .156 .016 1.242 .214 .979 1.021

PAGE 147

133 AWCPA .127 .032 .073 3.957 .000 .4 96 2.014 Grand Mesa Middle .019 .037 .009 .516 .606 .597 1.676 Franklin Middle .117 .034 .060 3.441 .001 .558 1.791 Ortega Middle .058 .034 .029 1.683 .092 .568 1.762 Colorado Springs East Middle 1.006 1.160 .405 .868 .386 .001 1287.548 Martin Luther King .047 .033 .171 1.444 .149 .012 82.594 Kunsmiller Middle .013 .031 .046 .411 .681 .013 74.643 Mt. Garfield Middle .140 .068 .507 2.056 .040 .003 359.201 Jefferson Middle .055 .057 .201 .964 .335 .004 256.280

PAGE 148

134 A linear regres sion was performed to determine if the following independent or predictor variables: gender, ethnicity, middle school, and participant in Gear UP or not could be used to estimate a student's knowledge about college entrance requirements Using these predic tor variables a significant model emerged (F=2.44, p<.000). Despite the significance of the predictor model the Adjusted R square = .005, suggests low predictability strength and the variables in the model only account for under 5% of the variance in the s urvey question answer. Several independent variables in the model were significant including: AWCPA Beta = .07 3, p=. 000 and Franklin Middle School Beta = 060 p=. 001 These significant independent variables are contributing to a portion of the variance i n the regression model. There does not seem to be any variable information on this particular question that merits much further thought or discussion, except that for whatever reason, two of the Colorado Springs middle schools did not appear as signi ficant here. Collinearity diagnostics were run to determine if there were high correlations amongst the predictor variables. Other than the three Colorado Springs middle schools, there was not a collinearity concern with any of the other variables. College entrance requirements 11 th grade. Table IV.4 4 Model Summary Research Question III Survey Question College Entrance Requirements 11 th Grade Model R R Square Adjusted R Square Std. Error of the Estimate 1 .157 a .025 .021 .351

PAGE 149

135 Table IV.4 5 ANOVA R esearch Question III Survey Question College Entrance Requirements 11 th Grade Model Sum of Squares df Mean Square F Sig. 1 Regression 18.089 21 .861 7.009 .000 a Residual 719.671 5856 .123 Total 737.760 5877

PAGE 150

136 Table IV.4 6 Coefficients Rese arch Question III Survey Question College Entrance Requirements 11 th Grade Model Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics B Std. Error Beta Tolerance VIF 1 (Constant) 1.939 .018 106.049 .000 GroupCod e .149 .016 .208 9.427 .000 .341 2.935 gender .030 .013 .031 2.404 .016 .987 1.013 American Indian .137 .054 .034 2.547 .011 .945 1.058 Asian .085 .043 .027 1.968 .049 .906 1.104 African American .080 .027 .044 2.968 .003 .753 1.329 H ispanic .077 .017 .106 4.622 .000 .317 3.150 Native Hawaiian .201 .095 .028 2.117 .034 .980 1.020 Not Reported .035 .062 .007 .570 .568 .976 1.024 White .086 .020 .072 4.360 .000 .604 1.656 Lamar HS .015 .023 .012 .663 .507 .496 2.014 AW CPA .041 .080 .007 .509 .611 .955 1.047 Martin Luther King .067 .027 .039 2.430 .015 .651 1.536 Aurora Central High .011 .023 .009 .485 .628 .537 1.863

PAGE 151

137 Grand Junction Central HS .017 .024 .012 .688 .492 .541 1.848 Northridge HS .025 .024 .019 1. 072 .284 .550 1.817 Greeley Central HS .051 .034 .022 1.497 .135 .778 1.285 Alamosa HS .021 .024 .016 .898 .369 .554 1.804 Wasson HS .018 .025 .012 .702 .483 .592 1.689 North HS .032 .025 .021 1.277 .202 .593 1.686 Montbello HS .066 .027 .040 2.480 .013 .630 1.588 Abraham Lincoln HS .055 .024 .040 2.325 .020 .552 1.810

PAGE 152

138 A linear regression was performed to determine if the following independent or predictor variables: gender, ethnicity, high school, and participant in Gear UP or not could be used to understand if a student had an understanding about graduation requirements. Using these predictor variables a significant model emerged (F=7.00, p=.000). Despite the significance of the predictor model the Adjusted R square = .021, suggests low pr edictability strength and the variables in the model only account for under 5% of the variance in the survey question answer. Several independent variables in the model were significant including: group Beta = .208, p=.000, gender Beta = .031, p=.016, Am erican Indian Beta = .034 p=.0 11 Asian Beta= .027, p= .049, African American Beta = .044 p=.00 3, Hispanic Beta = .1 06, p=.000 Hawaiian Beta = 02 8, p=.0 34 White Beta = .072, p=.000, Martin Luther King Beta = .39, p=.015, Montbello Beta= .040, p=.0 13, Lincoln Beta=.040, p=.020. These significant independent variables are contributing to a portion of the variance in the regression model. Holding all the other independent variables constant in the model, a typical CGU student is likely to answer .208 points more positively on the five point Likert scale than a student not in the CGU program that they understood college entrance requirements. This means if the control group mean score for this question is 1.88 the CGU student mean would be 2.088 The r egression here shows that CGU students felt as if they had an understanding of college entrance requirements much more strongly than those of their peers. It is possible that their peers, again as much as possible was done to match them up to the socioecon omic backgrounds of the CGU students, would not be hearing this message as often as needed because they may have been viewed as students who would not attend college.

PAGE 153

139 As with all the other survey questions utilized, CGU students answered more positively th an their peers. What is interesting here is that so many of the ethnic groups that were statistically significant, answered this question less positively than their peers. Unfortunately, any theories that might address this would be presumptive and it woul d be difficult to pinpoint exact reasons for this. Collinearity diagnostics were run to determine if there were high correlations am ongst the predictor variables. a tolerance level of .01 was used. Applying these criteria no variables fall below the .01 t olerance level therefore suggests there is not a multicollinearity issue with any of the variables CGU remedial course rates. Students may state they have knowledge of the courses they will need to complete in high school to be eligible for college, but the remedial class needs of CGU students classes. Table IV. 46 shows the CGU remediation rates for based on the high school they attended. Fifty five percent of all CGU students that attended college needed at least one remedial class upon enrollment, highlighting how they were underprepared academically to succeed in college, even though they may have known which classes they were expected to complete in high school. Fi fty percent of those students that enrolled in college remediation passed their class.

PAGE 154

140 Table IV.4 7 CGU Participant College Remediation Needs During First Year of College School Name Number of Students Assigned to Remediation i n at least one subject Number of Students Assigned to Remediation by Subject Math Writing Reading # % # # # Alamosa High School 63 29 46% 24 22 14 Aurora Central HS 80 59 73.80% 50 41 36 Central High School (Greeley) 147 70 47.60% 53 47 33 East High School (Pueblo) 83 50 60.20% 40 25 22 Lamar High School 26 12 46.20% 8 5 4 Lincoln High School 92 72 78.30% 64 54 47 MLK, Jr 31 18 58.10% 16 10 5 Montbello 95 76 80% 64 61 50 North High School 62 45 72.60% 43 36 31 Northridge High School 83 40 48.20% 32 28 26 Wasson High School 75 42 56% 28 31 17

PAGE 155

141 CHAPTER 5 DISCUSSION Introduction This chapter utilizes the data from the last chapter ascertained through the student survey for Research Questions I through III to explore the research q uestions posed in Chapter 1. Table V.1 highlights how CGU services studied in each research question had an effect on CGU participants. Table IV.1 Research Questions Sample Sizes and Beta Research Question Survey Question Sample Size 8th Grade Particip ants Sample Size 8th Grade Control Sample Size 11th Grade Participants Sample Size 11th Grade Control Beta 11 th Grade Test I Do you think that you will graduate from high school? 1387 90 1182 1396 .132 I Do you think that you will go on for fur ther education after you leave high school? 1385 90 1182 1394 .201 II What is your GEAR UP PCA doing for you? Rate how much you agree or disagree with the 1380 88 1172 1372 .302

PAGE 156

142 following statement: She/he will help me be able to go to college. III Ha s anyone from your school or GEAR UP ever spoken to you about Classes you need to take in high school to prepare for college? 1389 90 1182 1388 .157 III Has anyone from your school or GEAR UP ever spoken to you about College entrance requirements? 1389 90 1176 1380 .208 The hypotheses for Questions I through III are investigated and it is determined if CGU has an impact on student college aspirations, advisor/student relationships, college pathway knowledge, and if the use of a logic model had an impact on programmatic services. Study limitations describe potential shortcomings within the methodology, data collection, and analysis to inform the decisions of practitioners who many consider an effort to have this

PAGE 157

143 study inform future research on pre collegiate program implementation in general and CGU specifically, implications and suggestions for further research are provided, followed by the conclusion. Restatement of the Research Questions Research Question I: Is there a correlation between student participation in CGU and their educational aspirations? Hypothesis I.1: Low income students that participated in CGU will demonstrate higher aspirations to graduate from high school over time. H ypothesis I.2: Low income students that participated in CGU will demonstrate higher aspirations to earn a college degree over time. Research Question II: Is there a correlation between CGU PCA mentoring and student aspirations to pursue postsecondary edu cation? Hypothesis II.1: Low income students that participated in CGU will be more likely to pursue a postsecondary degree due to the strong relationship they developed with their PCA over time. Research Question III: Is there a correlation between stude nt participation in CGU and their knowledge and understanding of postsecondary pathways? Hypothesis III.1: Students aspirations to attend postsecondary education will increase over time. Research Question I : CGU Impact on Student Educational Aspirations. The statistical data used to determine what, if any impact CGU had on students that participated in the program vs. their peers who did not is a bit of a mixed bag. The statistical mean scores on high school and college aspirations do not appear to be ver y

PAGE 158

144 significant. Students who chose to participate in CGU had higher expectations to graduate high school and to attend college than their peers in both questions asked of students in the 8 th grade Mean scores increased for both sets of students for both qu estions in the 11 th grade. Although the average mean score increased slightly more for the students that participated in CGU, the increase was fairly small. While students in CGU surveyed very high in both questions, it could be inferred that these expecta tions were precisely why these students chose to participate in CGU in the first place. When regressions were utilized to break down the data by variables there were some very significant findings between CGU and their peer groups, but little to decipher with other variables. The statistics show that girls tested significantly higher than boys on both questions in both their 8 th grade year and their 11 th grade year of school That is consistent with other data that has been developed in other studies. The regressions showed a statistical significance as to high school and college aspirations between the CGU experimental group and the control group of their peers. As 8 th grade students there was no significance in the difference in answers between the two gr oups When the test was administered to the students in the 11 th grade those students that participated in CGU answered those questions more positively than their peers. It might be assumed that this is natural because the students who participated in CGU were more likely to answer positively by sure essence of their participation. But when you take into account that th grade, it gives credence to the effect CGU had. It should be noted however, that both gro ups, over the course of time felt strongly (an average close to 5: the top of the Likert scale) that they would graduate from high school and attend college.

PAGE 159

145 Unfortunately, the regressions do not provide telling data on other variables such as ethnicity or specific schools. This could be looked at in a couple of ways. It could be a sign that the strategies used by CGU were s ystemic enough that a higher or lower answer to either question by any individual ethnic group or school has little generalizability. T hat was one of the fundamental goals of the CGU program. Another way to look at this is to see this as a weakness of the study and to encourage future research that better breaks down these variables. One of the important goals of this study was to take ad vantage of the long term relationship between program and students to draw some lasting conclusions as to how CGU translated to student success (as measured by high school graduation, postsecondary matriculation, remediation needs, pass rates for students and student persistence in postsecondary education). When reviewing graduation rates, the data is not definitive either way. At first glance, the overall graduation rate for students who entered a CGU partner high school in 9 th grade (61.5%) is below the S tate average (75%) Two factors should be taken into account however. One is that average is consistent with the State average for students who are part of the same demographic (students on free reduced lunch.) Second, it might be reasonable to assume that there is a higher transient rate for these students and that they are more apt to leave their high schools than their peers. Adding the fact that CGU removed some students from the program due to the fact that they were undocumented students, the graduati on percentage of CGU students dropped and it would be hard to criticize the program for that. It is a weakness of this study that those numbers are not available.

PAGE 160

146 It is clear that a very high percentage of students who were in the program as 9 th grade stud ents, graduated from the program (over 90%). This is a substantial number any way one would choose to look at it. When trying to break down the overall data to specific variables (by ethnicity and by school) the data is inconclusive. The sample sizes are r elatively small. When broken down by school and ethnicity the change in responses is uneven and thus hard to generalize. When looking at how the students responded to the question on college asp irations, and how that translated, the conclusions to be draw n are similar. Both groups of students surveyed indicated strongly that they intended to go on to college in both their 8 th and 11 th grade years. Both numbers jumped a bit from 8 th to the 11 th grade. CGU average mean scores rose a bit more (.23 to .14) t han their peers. It would be hard to argue that the differences between the two groups were dramatic. The students who participated in CGU did have a significant positive rise in the 11 th grade in relation to their peers, but the regressions offer little i n the breakdowns by ethnicity or by school. A key indicator of the success of a pre collegiate program such as CGU should be student matriculation rates. Of those that graduated, 78% went on to postsecondary institutions. That compares favorably with the o verall percentage (67%) for the State of Colorado. When it is taken into account that the percentage of first time full time college freshmen were considered low income students ( 42 %) and of that only about (29%) received PELL that number is even more im pressive. The importance of the CGU scholarship, $1250 per semester, must be factored in however.

PAGE 161

147 Research Question II: CGU PCA Impact on Student Interest and Conviction in Postsecondary Education. This question was designed to focus on the relationship between the CGU student and their PCA. Specifically, how the CGU student perceived that their PCA would help them to matriculate to a postsecondary institution. The assumption that the students who did not participate in the CGU program and made up the con trol group would have a lower opinion of the relationship that they had with their counselors, based on the assumption that with the high student counselor ratios control students would not feel as strongly about their counselors ability to help them move forward to higher education. The analysis shows that there is a statistical difference in student perception for both groups in both the 8 th grade year and the 11 th grade year. In both years the CGU students felt more strongly about the ability of their a dvisor to help them than did the control students. What is interesting about the results is that the mean scores for each group dropped a bit in the 11 th grade from the mean scores of the 8 th grade. The dip is very small for the control group. The dip for the CGU experimental grip is larger and should raise some questions as to why. It is possible that even with all of the support from the PCAs, by the time CGU students reached the 11 th grade, factors such as student motivation, student performance, the ab ility of the family to support the students postsecondary goals (socially and economically) may have waned. Regardless, these are exactly the factors that the CGU is supposed to help students overcome. This dip should be a topic among CGU administration an d staff to determine strategies to address this outcome. The regressions utilized did not provide for much in the way of generalizable data.

PAGE 162

148 Girls again answered more positively then boys in both grades to the question. There was a noticeable jump in the positive response from CGU students to their peers. This should be seen as a realization of one of the main missions of CGU (a strong relationship with an adult). As important as it is for students to feel strongly that they have an adult that wi ll help them get into college, it is of greater importance that students have the knowledge and understanding to perform well upon matriculation. One way to measure this is to look at college retention numbers for students. These numbers should be monitor ed after every semester to help pre collegiate programs and the colleges understand where gaps are and why graduation rates are not better. It should be remembered that the PCAs stayed in contact with their students by having assigned office space and offi ce hours at many of the colleges and also used social media to help communicate with kids. Perhaps even more importantly, students receive their scholarship ($1250 per semester) and that would increase odds of having the financial support necessa ry to stay in school. For this study only retention rates for freshmen between their 1 st semester and their 2 nd semester are measured. When reviewing the figures, it is very clear that CGU had much better retention numbers then the students who were not p art of the program. These numbers may have been enhanced by the fact that CGU was able to track transfer students and in discussions with area colleges and universities, it was clear that not all were able to do so as quickly. In any case, it seems safe to say that CGU students were better able to adapt to all of the factors that matriculation to postsecondary institutions than their peers. Of particular importance is the disparity in retention rates at the area

PAGE 163

149 community colleges. One contributing factor m ight be that most of the dual enrollment that CGU students participated in occurred through area community colleges, which may have fostered an increased comfort level for CGU students. One other measure that might not have as much to do with the impact o f CGU is the pass rates of students at the college institutions. It would be beyond the scope of what to attribute success, or lack thereof, in college courses. Given that, the information gleaned from pass rates co uld be important for future research. Students in CGU did not fare as well as their peer counterparts. In discussions with the CGU administration and staff, it was their hope that offering dual enrollment courses while the student was still in high school would have better prepared the students for the rigor expected at the college level. For whatever reason, and any theories would just be speculative at this point, the pass rates do not match up well with their peers. There should be one point to track for further research. The pass rates are high for almost each school surveyed. That being the case, future studies should focus on what happens to these students throughout their college careers because these pass rates do not match up with the graduation rates at the colleges and universities. Research Question III: CGU Impact on Low Income Student Knowledge and Understanding of Postsecondary Pathways The question s utilized here was designed to address the issue of whether or not students were properly in formed of what courses students should take in high school to best prepare them for success in higher education. There is a concern across the education landscape that traditionally underserved students are not placed in courses that will

PAGE 164

150 properly prepare them for college. Traditionally these courses have been referred to as X courses, honors courses, or AP courses. Most recently the advent of dual enrollment courses has become part of this equation. Whatever the case may be, there is a strong belief that some students are not placed in rigorous classes in high school and that leaves them ill prepared to perform well in college. There is also the ongoing concern regarding the lack of alignment with curriculum between k 12 and higher education. This is most evident when students who have graduated from high school and been admitted into college are forced to retake classes in college covering content that was to be covered while the students were part of the k 12 system. These courses are sometimes referred to as developmental education but most of the time the word attached to this is remediation. Remediation is a growing concern for policy makers and educators across the country and it should be. Students who end up taking these courses in college have to p ay for them, receive no credit for passing them, and rarely graduate with a degree. The questions from the survey looked to address this and to help CGU bridge the knowledge gap for their students. The assumption would be that in the 8 th grade neither grou p would have heard too much about this either rigorous college preparation course or college entrance requirements Yet, the mean score of both the control and experimental group for both questions would indicate that they had. I will posit that if indeed they had, most of these students surveyed, CGU or not, could not describe either concept in any meaningful way. Moving forward, the assumption would be that more students had been educated on college pathways. The mean scores were relatively the same. The interpretation here

PAGE 165

151 could be looked at two ways. The first interpretation would be that all in K 12 are doing a very good job of outlining expectations to students about what courses are required to sufficiently prepare them for college. Another way to in terpret this data is to assume that students think they have a better grasp on the course pathways to college than they do. The regressions that were run to filter out variable specific data were largely inconclusive, with the exception of the 11 th grade students that participated in CGU answering in a significantly more positive way. One way to measure how these expectations are being absorbed by students is in remediation rates. The remediation rates for CGU students are extremely high, higher than the S tate average for Colorado (52% for a population of their peers). To be clear, CGU has a very limited ability to influence remediation rates directly. It may be a sign of the success of the CGU program that students with these deficiencies matriculated to c ollege and their retention rates are as good as they are. Still, evidence points to the fact that these students will have a difficult time graduating with a degree, and the high number of students needing remediation should be seen as a small weakness wit hin the CGU implementation. Study Limitations Every researcher enters their dissertation to add to the body of knowledge with previous individual perspectives, experiences and biases. As the Executive Director of CGU, I helped develop and coordinate all of the strategies that were utilized during the implementation of the grant. I have a deep personal passion for this work of helping traditionally underserved students have the sense that they have a realistic option to attend postsecondary institutions. I a m committed to the theory that the main difference

PAGE 166

152 in deciphering which students go to college and which do not is that those students who come from high socioeconomic backgrounds and have a history of attending college in their family do just fine. It is the students from low income families that struggle to attend college. It is the mission or pre collegiate programs like CGU to provide these underserved students the guidance, knowledge and motivation to believe that college is realistically available to them. Given my biases and personal investment in the program, providing quantifiable evidence was even more critical. Of the limited attempts to evaluate pre collegiate programs, most have centered on qualitative reviews. It was the goal of this work to pr ovide numerical data to the body of knowledge surrounding pre collegiate programs in general, and GEAR UP specifically. Additional limitations include the number of variables in the study and research models. Although a number of the models are statistical ly significant in predicting the outcome variable or answer to various survey questions the relatively low predictive power as suggested by the R squared numbers suggest that there are other variables at play that have an influence on how the students may have answered the questions. As with any research, there are always additional variables that may have strengthened the research models if it would have been possible to co llect and quantify such factors. Implications The findings of this study lead to so me direct implications that could influence practioners as they plan to implement similar programs. First, pre collegiate programs should have a specific curriculum that addresses all pertinent information to help students navigate their way through high s chool and best be prepared to take advantage of all of

PAGE 167

153 the different level of opportunities at postsecondary institutions. It is clear that the amount of time a school counselor can spend with each individual student is not sufficient. This is not the faul t of counselors. It is a broken concept that if modeled after CGU, could produce significantly improved results. career as possible should be exhausted. Students began takin g college classes as early as the second semester of their sophomore year as part of CGU. The overall success rate (as planning and evaluation must take place to properly place a student in a college class, but the message that college is the expectation is delivered in a much more meaningful way than has been offered up before. Third, alternative methods of instruction should be explored. With the increasing reliance on techno logy and its ability to adapt more specifically to individual student needs and learning styles, online education is most certainly going to become a more standard method of delivery. It would behoove all practitioners, to acknowledge this and to strategiz e how it can best aid their students. Fourth, remediation must be addressed in a meaningful way that produces significant results. Tweaking at the margins in terms of instruction, or minor alterations to curriculum will not suffice. Significant alternativ e strategies must be developed. In the new CGU grant, remediation is being addressed in middle schools. By simply aligning country can stay clear of these courses. In many cases, it is a matter of just renaming existing high school courses.

PAGE 168

154 Fifth, alternative methods of earning college credit should continue to be utilized. test provided CGU students with almost 5,000 col lege credits. Almost 89% of students (some students were as young as freshmen in high school) earned college credits through this 75 minute test. Finally, pre collegiate programs must begin to be measured on their return on investment. Millions of dollars have been spent on pre collegiate programs over the years and the access and achievement gap has only increased. There must be more accountability. A cost benefit analysis of the GEAR UP program takes into account the total costs associated with running th e program over 6 years, the number of participants that will degree. First Table V.2 shows the full number of enrollments through Table V. 2 Total Enrollments Based on Years in CGU Cohort Students Years in Program # of Enrollments Cohort 1 1,200 7 8,400 Cohort 2 1,200 6 7,200 Cohort 3 1,20 0 5 6,000 Cohort 4 1,200 4 4,800 Total # of Enrollments Throughout Program Life Cycle 26,400 As shown in Table V.3 c osts are fixed at the level the grant was funded at which was $21 million over 6 years. $10.5 million was spent on operating costs whi ch included salaries and benefits, materials and supplies, concurrent enrollment tuition and fees and

PAGE 169

155 charges for all other programmatic activities. The other $10.5 million was spent on college scholarships for participants, with each student receiving up to $10,000. Table V.3 Cost p er Enrollment in CGU Funds Dollars Per Fund $ Divided by Enrollments Operating Funds $10,500,000.00 $397.73 Scholarship $10,500,000.00 Total $21,000,000.00 $795.45 The economic benefits of the GEAR UP program can best be measured by the additional taxes that their additional earnings will generate over their lifetime in the work force. Table V.4 highli ghts these findings. Studies have shown that the taxpayer benefit from high state and federal income taxes paid on the higher salaries earned by college graduates varies from $60,000 to $150,000 in additional income taxes paid over the work life of the gra duate, depending on the selectivity of the college attended. For this cost benefit analysis, the conservative estimate of $60,000 in lifetime taxes paid was used. The analysis also estimates conservatively that 42% of GEAR UP students will graduate from This means that the GEAR UP participants that taxes over their work life. Subtracting the total cost for the GEAR UP program and calcula ting a present value of these payments using a 7% discount rate over an estimated 40 year career span results in the benefit of the program equaling around $6.6 million in Thes e

PAGE 170

156 means the federal government receives nearly a 2 for 1 return on their investment in the Colorado GEAR UP program. Table V.4 Economic Cost Benefit Analysis of CGU Lifetime Taxes Paid by Bachelor's Degree Holder $60,000 Estimated GEAR UP Bachelor degree attainment of 42% X 2,000 Additional Taxes Paid by GEAR UP students Over Lifetime = $120,000,000 Less Cost of GEAR UP Program $21,000,000 Total Benefit of GEAR UP Program Over 40 Years = $ 99,000,00 0 Present Value of Program's Lifetime Benefit Less Costs (Discount Rate of 7%, 40 years of Earnings) $ 6,611,258 Suggestions for Further Research This study explored low program and how expos ure to information regarding the importance of earning a college degree changed their biases, goals, and behaviors toward postsecondary education. I college in an effort to discover if college knowledge/readiness programs such as CGU framing the study leave ample room for additional research related to student outcomes after compl etion of college knowledge/readiness curriculum. As the program continues, a longitudinal study could illuminate the impact CGU beyond the successes of study parti cipants, it would be particularly useful to know how postsecondary aspirations. As the largest portion of the research participants came from

PAGE 171

157 Hispanic backgrounds, it would also be useful to explore how students from other ethnicities perceived their involvement influenced their transition, persistence, and progression through college. Similarly, are there any relationships among familial income levels and student goal s for postsecondary education? While all these students come from low income families, does their placement in the spectrum of income levels matter? their developmen t, others also involved in their learning may provide insights that students do not perceive. Parents and siblings, friends, high school teachers, and college instructors could enhance the understanding of how college knowledge/readiness curriculum affects education. These student outcomes are influenced by programmatic choices made within the CGU, and further research could investigate services and interventions provided through other state wide GEAR UP programs as well as other college readiness programs. Exploration of those models could lead to best practices to be shared among programs, and could establish a body of evidence for the effectiveness and necessity of such programs. Additional ly, how do students from similar backgrounds perceive their collegiate opportunities, and how to those views compare to students in college perception about college, but the addition of a control group would be useful to gauge the how large those changes were in comparison to student who were not exposed to college knowledge/readiness curriculum.

PAGE 172

158 Finally, how do these programs benefit high schools and universities? Do they impact community colleges? Exploration into how college knowledge/readiness attainment, and career choices would be beneficial. Interviews with university administrat ors could discover broader intervention opportunities for students on campus designed to continue the services they receive while in high school, and interviews with governmental decision makers could highlight outcomes and evidence that would influence th eir likelihood of funding similar initiatives. Learning from the Dissertation Process There are some weaknesses to this study that are correctable for future studies. In ascending order of importance they are: an emphasis should be placed on motivation t heory. It is my contention that when working with students from low income backgrounds who would be the first in their family to attend college, that it is not enough to know the material; in this case pathways to college curriculum. While that is importan t, it is only a small part of the work. The major hurdle is getting these students to believe that college is truly available and attainable and then creating that expectation. The second adjustment that could be made is the use of a much better instrument to gauge student learning on college pathways than the one that I used. Third, I do not believe that the role that PCAs play in the success of CGU is properly illuminated. They are invaluable. In my opinion, student motivation is the key factor in predict ing a success in matriculating from high school to college. Motivation can be measured in different ways. But many could predict to a pretty high degree which kids from any high

PAGE 173

159 school will most likely matriculate to college. I maintain that stud ents who come from high socioeconomic families are going to college, regardless of anything that might be done at any individual school. This is not to discount the good work of principals, counselors and teachers at these schools. Still, the reality is th at those kids have had the expectation of attending college from the beginning of their lives and they know that they have the means to get there. I do not see any great trick in getting students from those family backgrounds to college. The real trick is getting underrepresented students to college and motivation is central to that task. Students who think they are going to college will do what is necessary to get there. They will go to class, pay attention, do the homework, study for tests, and get good g college do any of those things beyond the bare minimum amount of effort? enough about motivation theory to say much, but I can assure you that motivating kids is central to this w ork. This ties in to my second point as to the value of the PCAs. One of the weaknesses of my study is that as hard as I tried to pinpoint the strongest PCAs from the es that are done, the importance of a strong adult who has the knowledge and the ability to motivate a student, will come through. Which leads to my final point ; I purposefully stayed away from interviews and qualitative data. There are not enough studies done on pore collegiate programs and the studies that are done are lacking in accountable data. However, there is information to be

PAGE 174

160 captured in the data. A mixed methods stu dy could help bridge some of these gaps. Interviews with both students and PCAs could add a great deal in the telling of these stories. Conclusion Additionally, jobs within man y industries are being outsourced overseas, and professions higher cost of living. Education leaders and leaders of pre collegiate programs must find options to bridge t he achievement gap and the attainment gap for low income students in this country. One of the main goals of this study was to provide quantifiable data to pre collegiate programs. When reviewing the data by survey question, by school, or by ethnicity it is inconclusive. However, the bottom line for any pre collegiate program such as CGU should be college matriculation rates. The bottom line for CGU was that 78% of their seniors went on to college. That is higher than the overall matriculation for the studen ts in the State of Colorado (67%) by a pretty substantial margin. That number is even more significant when reviewing these matriculation figures for low income students. According to the Colorado Department of Higher Education, only 42% of the first time, full time freshman were students from low income backgrounds. Only 29% of those students applied for PELL federal dollars. As significant as those numbers appear to be, it can be argued that there are many other variables to consider when looking at the scope of pre collegiate programs

PAGE 175

1 61 throughout the country. It is my hope that if nothing else, this study helps provide some baseline quantitative data for future programs to utilize. This study provided a window to view one program that could provide strat egies to help achieve this goal. These strategies are formulaic and can produce similar results in all states across the country. This is not an advocating for increased funding. It is a call to utilize funding in a much more systemic and systematic way th at calls for metric outcomes for all programs of its like.

PAGE 176

162 References Adelman, C. (2006). The toolbox revisited: Paths to degree completion from high school through college Washington, DC: U.S. Department of Education. Alexitch, L. R., Kobussen, G. P., & Stookey, S. (2004). High School Students' Decisions to Pursue University: What Do (Should) Guidance Counsellors and Teachers Tell Them? Guidance & Counseling, 19 (4), 142 152. American Council on Education (2009). Minorities in higher education: 200 9 supplement. Washington DC: Mikyung Ryu. Bailey, T. R., Hughes, K. L., & Karp, M. M. (2002). What role can dual enrollment programs play in easing the transition between high school and postsecondary education? Washington D.C.: Office of Vocational and Adult Education, U. S. Department of Education. Bailey, T. R., Hughes, K. L., & Karp, M. M. (2003). Dual enrollment programs: Easing transitions from high school to college. New York: Community College Research Center, Teachers College, Columbia Univer sity. Collaborating for student success. Education, 125 (2), 259 270. Blanchard, B. E. (1971). A national survey of curriculum articulation between college of

PAGE 177

163 liberal arts and s econdary school. Chicago: De Paul University. Bruce, A. M., Getch, Y. Q., & Ziomek Daigle, J. (2009). Closing the Gap: A Group Counseling Approach to Improve Test Performance of African American Students. Professional School Counseling, 12 (6), 450 457. C. S. Mott Committee on After School Research and Practice. (2005). Moving toward success: Framework for after school programs. Washington, DC: Collaborative Communications Group. Cabrera, A. F., Deil Amen, R., Prabhu, R., Terenzini, P. T., Chul, L., & Franklin Jr, R. E. (2006). Increasing the College Preparedness of At Risk Students. Journal of Latinos & Education, 5 (2), 79 97. doi: 10.1207/s1532771xjle0502 2 \ Cabrera, A. F., & La Nasa, S. M. (2000). Understanding the college choice process. New Di rections for Institutional Research, 2000 (107), 5. Camizzi, E., Clark, M. A., Yacco, S., & Goodman, W. (2009). Becoming "Difference Makers": School University Collaboration to Create, Implement, and Evaluate Data Driven Counseling Interventions. Professi onal School Counseling, 12 (6), 471 479

PAGE 178

164 Castillo, L. G., Conoley, C. W., Cepeda, L. M., Ivy, K. K., & Archuleta, D. J. (2010). Mexican American Adolescents' Perceptions of a Pro College Culture. Journal of Hispanic Higher Education, 9 (1), 61 72. doi: 10.1 177/1538192709350454 Colorado Department of Higher Education. The impact of public higher education on the state of colorado (2007). Retrieved from http://highered.colorado.gov/Publications/Studies/2007/200712 ImpactofHE.pdf Conley, D. T. (2007). Tow ard a more comprehensive conception of college readiness. Eugene, OR: Educational Policy Improvement Center. Constantine, M., Kindaichi,M., & Miville, M. (2007). Factors influencing the educational and vocational transitions of Black and Latino high sch ool students. Professional School Counseling, 10 (3), 261 265. Cooper, R., & Liou, D. D. (2007). The structure and culture of information pathways: Rethinking Opportunity to learn in urban high schools during the ninth grade transition. High School Journ al, 91 (1), 43 56. De La Rosa, M., & Tierney, W. (2006). Breaking through barriers to college: Empowering low income communities, schools, and families for college opportunity and student financial aid Los Angeles, CA: Center for Higher Education Policy Analysis.

PAGE 179

165 Downs, A., Martin, J., Fossum, M., Martinez, S., Solorio, M., & Martinez, H. (2008). Parents Teaching Parents: A Career and College Knowledge Program for Latino Families. Journal of Latinos & Education, 7 (3), 227 240. doi: 10.1080/1534843080210 0295 Plans. School Counselor 44, 384 394. Fuller, H. (2002). Educational choice, a core freedom. The Journal of Negro Education, 71 (1/2), 1 4. Gall, M.D., W.R., & Gall, J.P. (1996). Educational Research: An introduction White Plains, NY: Longman Publishers. Gandara, P. & Bial, D. (2001). Paving the way to post secondary education: K 12 intervention programs for underrepresented youth (NCES 2001 205). Washingto n D.C. National Postsecondary Education Cooperative Access Working Group, U.S. Department of Education, National Center for Education Statistics. Garton, G. (2003). Dual enrollment: Are students truly successful and what ensures their

PAGE 180

166 success? (Unpublish Missouri). Garza, E., Barnett, B. Merchant,B., Soho,A. &Smith, P. (2006). The Urban School Leaders Collaborative: A school university partnership emphasizing instructional leadership and student a nd community assets. International Journal of Urban Education 1, 14 30. Gibson, D. M., & Jefferson, R. N. (2006). The effect of perceived parental involvement and the use of growth fostering relationships on self concept in adolescents participating in G EAR UP. Adolescence, 41 (161), 111 125. Gose, Ben (2006). Colorado Debates How to Send More At Risk Students to College. The ChroniclE of Higher Education. May, 2006. Greenberg, A. R. (1989). Concurrent enrollment programs: College credit for high scho ol students. Bloomington, Indiana: Phi Delta Kappa Educational Foundation. Greene, J. P., & Forster, G. (2003). Public high school graduation and college readiness rates in the United States New York: Manhattan Institute. Grubb. W. N. (1999). Learnin g and earning in the middle: The economic benefits of sub

PAGE 181

167 baccalaureate education New York: Community College Research Center, Columbia University. Harvard Family Research Project. (1999). Learning from logic models: An example of a family/school partn ership program. Cambridge, MA. Hewett, S. M., & Rodgers, W. J. (2003). The Citadel GEAR UP program and learner centered education: Together, a framework for student success. Education 124 (1), 86 91. House, Gerry N. (2008). Exploring the Race and the Gend er Gaps National Opportunity to Learn Summit: Building a Philanthropic Movement for Systemic Change. October 2008. Janson, C. (2009). High school counselors' views of their leadership behaviors: A qualitative methodology study." Professional School Coun seling, 13 (2), 86 97. White SAT gap is actually Growing larger. The Journal of Blacks in Higher Education, (25), 95 100. Kahlenberg, R.D. (2006). Integration by income. American School Board Journal, 193 (4), 51 52.

PAGE 182

168 Kirst, M., & Venezia, A. (2001). Bridging the great divide between secondary schools and postsecondary education. Phi Delta Kappan, 83 (1), 92 98. Kozol, J. (2001). Ordinary resurrections. Children in the years of hope New York: Harper C ollins. Krei, M. S., & Rosenbaum, J. E. (2001). Career and college advice to the forgotten half: What do counselors and vocational teachers advise?" Teachers College Record 103 (5), 823. Lee, L. J., & Sampson, J. F. (1990). A practical approach to progra m evaluation. Evaluation and Program Planning, 13, 157 164. Lozano, A., Watt, K. M., & Huerta, J. (2009). A comparison study of 12th grade Hispanic students' college anticipations, aspirations, and college preparatory measures. American Secondary Educat ion, 38 (1), 92 110. Marsico, M. & Getch, Y. Q. (2009). Transitioning Hispanic seniors from high school to college. Professional School Counseling, 12(6), 458 463. McCarron, G. & Inkelas, K. (2006). The gap between educational aspirations and Attainment for first generation college students and the role of parental Involvement. Journal of College Student Development, 47 (5), 534 549.

PAGE 183

169 McElvain, C. C., Judith, G., & Diedrich, K. C. (2006). Beyond the bell: Tools for school o effective after school programs. Naperville, IL: Learning Point Associates. National Association of Colleges and Employers, Spring 2011 Salary Survey. (2011). From http://www.naceweb.org/research/salary survey/?referal=research&menuID=71. National Ce nter for Education Statistics (NCES). (2002). The condition of education 2002. Washington, D.C. U.S. Department of Education. National Center For Public Policy and Higher Education (2008). Measuring up: The national report card on higher education. San Jose, CA. Strengthening college preparation and access through concurrent enrollment in high school and community college. Austin, TX: University of Texas at Austin. Nunez, A., & Cuccaro Alamin, K., (1998). Fir st generation students: Undergraduates Whose parents never enrolled in postsecondary education (NCES 98 082).

PAGE 184

170 Washington DC: U.S. Department of Education.Orr, M. T. (1998). Integrating secondary schools and community colleges through school to work transit ion and education reform. Journal of Vocational Education Research, 23 (2), 93 113. Orr, M. T. (1998). Integrating secondary schools and community colleges through school to work transition and education reform. Journal of Vocational Education Research, 2 3 (2), 93 113. Orr, M. T. (1999). Community college and secondary school collaboration of workforce development and education reform. A close look at four community colleges. New York: Community College Research Center, Teachers College, Columbia Universi ty. Orr, M. T. (2002). Dual enrollment: Developments, trends and impacts. New York: Community College Research Center, Teachers College, Columbia University. Osborn, J. W. (1928). Overlapping and omission in our course of study. Bloomington, IL: Publ ic School Publish Company. Patton, M. Q. (1997). Utilization focused evaluation: The new century text. Thousand Oaks, CA: Sage Publications. Pennington, H. (2004). Fast track to college: Increasing postsecondary success for all

PAGE 185

171 students. Renewing our sc hools, securing our future: A national task force on public education Boston, MA: Jobs for the Future. (ERIC Document Reproduction Service No. ED486158) Perna, L. (2006). Understanding the relationship between information about college Prices and financi related behaviors. American Behavioral Scientist, 49 (2), 1620 1635. Postsecondary Education Opportunity (2008). College Participation Rates for Students from Low Income Families by State FY1993 toFY2006. February 2008. Issue Number 188. Rosenbaum, J. E. (1998). Unrealistic plans and misdirected efforts: Are community colleges getting the right message to high school students? New York: Community College Research Center, Teachers College, Columbia University. Royse, D., Thye r, B. A., Padgett, D. K., & Logan, T. K. (2001). Program evaluation: An introduction (3 rd edition). Belmont, CA: Brooks/Cole Thomson Learning. Watanabe, Teresa (2008). Lack of skilled workers will lead to fiscal crisis, experts say Los Angeles Times (20 08).

PAGE 186

172 W. K. Kellogg Foundation. (2004). Using logic models to bring together planning, evaluation, and action. Battle Creek, MI: Author. Zusman, A. (1999). Issues facing higher education in the twenty first Century. In P.G. Yazzie Mintz, E. (2010). Char ting the path from engagement to achievement: A report on the 2009 high school survey of student engagement Bloomington, IN: Center for Evaluation & Education Policy.

PAGE 187

173 Appendix Survey Questions 1. What is your current grade level? (Select your answer) 2. Do you think that you will graduate from high school? (Select your answer) 3. Do you think that you will go on for further education after you graduate from high school? (Select your answer) 4. What is the highest level of education that you think you will achieve? (Ma rk one answer that best describes how you feel now). 5. Has anyone from your school or GEAR UP ever spoken to you about the following: Credits you need to earn to graduate from high school 6. Has anyone from your school or GEAR UP ever spoken to you about the fo llowing: Classes you need to take in high school to prepare for college 7. Has anyone from your school or GEAR UP ever spoken to you about the following: College entrance requirements 8. Has anyone from your school or GEAR UP ever spoken to you about the followi ng: Availability of financial aid to help you and your family pay for you to go to college 9. What is your GEAR UP doing for you? Rate how much you agree or disagree with the following statements: It's helping me get better grades in school 10. What is your GEAR UP PCA doing for you? Rate how much you agree or disagree with the following statement: She/he will help me be able to go to college.

PAGE 188

174 11. What is your GEAR UP PCA doing for you? Rate how much you agree or disagree with the following statements: It has convin ced me to go to college 12. What part of your GEAR UP has influenced you the most to this point? Rate how much the following parts of the program have influenced you: GEAR UP workshops/activities 13. What part of your GEAR UP has influenced you the most to this p oint? Rate how much the following parts of the program have influenced you: One on one time with your coordinator 14. What part of your GEAR UP has influenced you the most to this point? Rate how much the following parts of the program have influenced you: Men toring 15. What part of your GEAR UP has influenced you the most to this point? Rate how much the following parts of the program have influenced you: Tutoring or help with schoolwork 16. What part of your GEAR UP has influenced you the most to this point? Rate how much the following parts of the program have influenced you: Visits to college campus 17. What part of your GEAR UP has influenced you the most to this point? Rate how much the following parts of the program have influenced you: Information about financial ai d and how much college costs 18. What part of your GEAR UP has influenced you the most to this point? Rate how much the following parts of the program have influenced you: Information about the benefits of going to college

PAGE 189

175 19. What part of your GEAR UP has influen ced you the most to this point? Rate how much the following parts of the program have influenced you: College scholarship 20. What part of your GEAR UP has influenced you the most to this point? Rate how much the following parts of the program have influenced you: The GEAR UP coordinator in my school 21. Has anyone from your school or GEAR UP ever spoken to you about the following: Financial aid 22. Has anyone from your school or GEAR UP ever spoken to you about the following: Grants 23. Has anyone from your school or GEAR UP ever spoken to you about the following: Loans 24. Has anyone from your school or GEAR UP ever spoken to you about the following: Scholarships 25. Has anyone from your school or GEAR UP ever spoken to you about the following: Work study 26. Do you think that you co uld afford to attend college or a vocational school using financial aid, scholarships, and your family's resources? (Select your answer) 27. Has anyone from your school or GEAR UP ever spoken to you about the following tests: EXPLORE 28. Has anyone from your schoo l or GEAR UP ever spoken to you about the following tests: ACT

PAGE 190

176 29. Has anyone from your school or GEAR UP ever spoken to you about the following tests: PSAT 30. Has anyone from your school or GEAR UP ever spoken to you about the following tests: SAT 31. Has anyone fro m your school or GEAR UP ever spoken to you about the following tests: Accuplacer 32. In high school, the college entrance requirements require that I take: ENGLISH 33. In high school, the college entrance requirements require that I take: MATH 34. In high school, the college entrance requirements require that I take: SCIENCE 35. In high school, the college entrance requirements require that I take: FOREIGN LANGUAGE 36. Has anyone from your school or GEAR UP ever spoken to you about the following: Vocational/Technical certific ate program 37. Has anyone from your school or GEAR UP ever spoken to you about the following: Associate's degree (2 year) 38. Has anyone from your school or GEAR UP ever spoken to you about the following: Bachelor's degree (4 year) 39. Has anyone from your school or GEAR UP ever spoken to you about the following: Master's degree (2 more years beyond a bachelor's degree) 40. Has anyone from your school or GEAR UP ever spoken to you about the following: Doctoral degree (4 more years beyond a bachelor's degree)

PAGE 191

177 41. Since the beg inning of the school year, how often have you talked with either or both of your parent(s) or guardian(s) about the following: School activities or events of interest to you 42. Since the beginning of the school year, how often have you talked with either or b oth of your parent(s) or guardian(s) about the following: Things you studied in class 43. Since the beginning of the school year, how often have you talked with either or both of your parent(s) or guardian(s) about the following: Your grades 44. Since the beginnin g of the school year, how often have you talked with either or both of your parent(s) or guardian(s) about the following: Preparing for high school 45. Since the beginning of the school year, how often have you talked with either or both of your parent(s) or g uardian(s) about the following: Going to college 46. What is the highest level of education that you think your parent(s) or guardian(s) want you to achieve? 47. Rate how much you agree or disagree with the following statements concerning your parent(s) or guardia n(s): My parents told me they expect me to graduate from high school 48. Rate how much you agree or disagree with the following statements concerning your parent(s) or guardian(s): My parents expect me to go to college

PAGE 192

178 49. Rate how much you agree or disagree with the following statements concerning your parent(s) or guardian(s): My parents encourage me to get good 50. What is the main reason you would not continue your education?

PAGE 193

179 51. FOR HIGH SCHOOL SENIORS ONLY Have you applied to any colleges for next year?