Citation
The Relationship of High-Risk Courses to First Year Drop-Outs at a Small Rural Community College

Material Information

Title:
The Relationship of High-Risk Courses to First Year Drop-Outs at a Small Rural Community College
Creator:
Lay, Terolyn G
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (104 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ed.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Higher Education Administration
Human Development and Organizational Studies in Education
Committee Chair:
CAMPBELL,DALE FRANKLIN
Committee Co-Chair:
KRAMER,DENNIS ALLAN
Committee Members:
MCFARLIN,ISAAC
GAGE,NICHOLAS A

Subjects

Subjects / Keywords:
academic -- college -- community -- momentum -- retention -- rural
Human Development and Organizational Studies in Education -- Dissertations, Academic -- UF
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Higher Education Administration thesis, Ed.D.

Notes

Abstract:
Student attrition, or dropping-out, from college is a source of concern as it impacts not only the student but the institution as well. Current literature, while rich in student retention studies at four-year institutions, provides little insight into factors impacting student retention at rural community colleges. This study, guided by the framework of academic momentum, is intended to fill a gap in rural community college literature. This study evaluated high-risk courses, in correlation with selected demographic and academic variables, for their contributions to first-year student drop-out risk. Data was collected to assess a cohort of first-time-in-college, associate-of-arts or associate-of-science degree seeking students attending a small, rural community college located in the panhandle of Florida. Regression analyses were utilized to understand the effect of high-risk-course-taking behavior, in conjunction with other student attributes, on retention rates after one-semester and one-year of attendance. Additionally, the effect of high-risk courses on college GPA as a measure of student success was examined as was the effect of credit accrual. Results showed that high-risk courses, apart from Intermediate Algebra, were not a factor in retention rates. High-risk courses were a factor in student success as measured by college GPA. Successful accrual of credit was found to be the best measure of student success and retention. The results of this study provide a deeper understanding of factors associated with first-year drop-out risk which can assist in advising and monitoring poor-performing students before they exit the institution. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ed.D.)--University of Florida, 2017.
Local:
Adviser: CAMPBELL,DALE FRANKLIN.
Local:
Co-adviser: KRAMER,DENNIS ALLAN.
Statement of Responsibility:
by Terolyn G Lay.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
LD1780 2017 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

THE RELATIONSHIP OF HIGH RISK COURSES TO FIRST YEAR DROP OUT S AT A SMALL RURAL COMMUNITY COLLEGE By TEROLYN LAY A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMEN TS FOR THE DEGREE OF DOCTOR OF EDUCATION UNIVERSITY OF FLORIDA 2017

PAGE 2

2017 T e rolyn Lay

PAGE 3

To Richard

PAGE 4

4 ACKNOWLEDGMENTS The decision to pursue a doctorate was easy. The ability to stay in the program and finish would have been impossible without the encouragement and assistance of so many individuals. I will forever be grateful for their unflaggin g support. First, I would like to say thank you to Dr. Dale F. Campbell, my chair and the pressure was high. I would also like to thank my committee members, Dr. Dennis K ramer, Dr. Isaac McFarlin, and Dr. Nicholas Gage, for their service and encouragement during this process. Next, to the 2014 LEAD cohort, I cannot begin to describe how appreciative I am of each of you. Having shoulders to lean on, even virtually, made th is process so much easier. I especially owe a debt of gratitude to Will Hamilton whose knowledge of STATA saved me during the data analysis I owe special thanks to Cris Lloyd and Bee Gudapati for being my soul sisters. Whenever the stress was high, I coul d always count on you to provide a ray of hope and produce a smile on my face. I would not have been able to pursue this degree without the assistance and encouragement from my colleagues at Chipola College. T he administration, faculty and staff have given me encourag ement and pep talks over the last four years. To the members of the Mathematics and Natural Sciences department I owe extra special thanks. Without your willingness to help in whatever way possible I would have been unable to successful ly navigate the duties at Chipola and the requirements of the doctoral program. Lastly, and most importantly, thank you to my family. To my daughters, Rachel and Nicole, thank you for listening, for reminding me of my bucket list, and most of all,

PAGE 5

5 for lovi ng me. To my step daughters, Laura, Melanie, and Ashley, thank you for all the words of encouragement and love. Finally, to my husband Richard, without your unwavering support I would have never successfully completed my doctorate. You are

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLE DGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ ........ 10 LIST OF ABBR EVIATIONS ................................ ................................ ........................... 11 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 Purpose Statement ................................ ................................ ................................ 17 Research Questions ................................ ................................ ............................... 18 Significan ce of the Study ................................ ................................ ........................ 19 Organization of the Study ................................ ................................ ....................... 19 2 LITERATURE REVIEW ................................ ................................ .......................... 21 Reasons for Dropping Out ................................ ................................ ...................... 21 Benchmarks for Success ................................ ................................ ........................ 23 High School Grade Point Average (GPA) ................................ ......................... 24 Gender ................................ ................................ ................................ ............. 25 Race/Ethnicity ................................ ................................ ................................ .. 26 Pell Status ................................ ................................ ................................ ........ 26 Full time vs. Part time Status ................................ ................................ ........... 27 First semester and First year College GPA ................................ ...................... 28 Gateway/Gatekeeper Courses ................................ ................................ ......... 29 The Rural Community College ................................ ................................ ................ 31 Theoretical/Conceptual Frameworks ................................ ................................ ...... 33 Historical Frameworks ................................ ................................ ...................... 33 Academic Momentum Framework ................................ ................................ .... 35 Summary of Existing Research ................................ ................................ ............... 36 3 METHODOLOGY ................................ ................................ ................................ ... 38 Research Questions ................................ ................................ ............................... 38 Setting ................................ ................................ ................................ ..................... 39 Data Collection ................................ ................................ ................................ ....... 40 Sample Population ................................ ................................ ........................... 41 High Risk Courses ................................ ................................ ........................... 41 Measures ................................ ................................ ................................ ................ 42

PAGE 7

7 Dependent Variables ................................ ................................ ........................ 42 Independent Variables ................................ ................................ ..................... 43 Covariates ................................ ................................ ................................ ........ 43 Analysis Methods ................................ ................................ ................................ .... 44 Limitations of the Study ................................ ................................ ........................... 47 Summ ary of Methodology ................................ ................................ ....................... 47 4 RESULTS ................................ ................................ ................................ ............... 51 Descriptive Statistics: Gender, Ethnicity, Enrollment Status, Pell Status, HS GPA and High Risk Course Enrollment ................................ ............................... 51 Research Question 1: Effect of High Risk Courses and other Factors on Fall to Spring Retention ................................ ................................ ................................ .. 52 Research Question 2: Effect of High Risk Courses and Other Factors on Fall to Fall Retention ................................ ................................ ................................ ...... 53 Fall to Fall Retention: Entering Cohort ................................ ............................. 53 Fall to Fall Retention: Students with Continuous Enrollment ............................ 54 Research Question 3: First Year College GPA and High Risk Courses as a Measure of Academic Success ................................ ................................ ........... 54 Additional Findings ................................ ................................ ................................ 55 Summ ary Of Results ................................ ................................ ............................... 57 5 DISCUSSION AND CONCLUSION ................................ ................................ ........ 67 Purpose of Study Reviewed ................................ ................................ .................... 67 Di scussion of Results ................................ ................................ .............................. 68 Significant Characteristics of Drop Out Risk ................................ ........................... 70 Part time vs Full time Enrollment ................................ ................................ ..... 70 Pell Status ................................ ................................ ................................ ........ 71 Gender ................................ ................................ ................................ ............. 72 Ethnicity ................................ ................................ ................................ ............ 73 Academic Preparedness via HS GPA and Retention ................................ ....... 73 Academic success via college GPA and Retention ................................ .......... 74 Accruing Credit ................................ ................................ ................................ 75 Implicatio ns for Institutional Practice ................................ ................................ ....... 76 Recommendations for Further Research ................................ ................................ 81 Limitations ................................ ................................ ................................ ............... 82 Summary of Discussion ................................ ................................ .......................... 83 APPENDIX A ADDITIONAL RESULTS ................................ ................................ ......................... 85 B INSTITUTIONAL IRB APPROVALS ................................ ................................ ....... 89 LIST OF REFERENCES ................................ ................................ ............................... 91 BIOGRAPHIC AL SKETCH ................................ ................................ .......................... 104

PAGE 8

8 LIST OF TABLES Table page 3 1 Ethnic representation of Florida and its state colleges ................................ ....... 48 3 2 Summary table of dependent variables ................................ .............................. 48 3 3 Summary table of independent variables ................................ ........................... 49 3 4 Summary table of covariates ................................ ................................ .............. 49 3 5 Risk ratio for students enrolled during Fall 2013 ................................ ................. 49 3 6 Cumulative risk ratio for students enrolled both Fall 2013 and Spring 2014 ....... 49 4 1 Descriptive statistics: Fall 2013 cohort ................................ ............................... 58 4 2 First semester high risk course taking behavior ................................ ................. 58 4 3 First year high risk course taking behavior ................................ ......................... 58 4 4 Effect of MAT1033 on Fall to Spring retention ................................ .................... 59 4 5 Logistic regression analysis for Fall to Spring retention ................................ ..... 59 4 6 Logistic regression analysis for Fall to Spring retention with inclusion of HS GPA ................................ ................................ ................................ .................... 59 4 7 Logistic regression analysis for likelihood of Fall to Fall retention after one semester of attendance ................................ ................................ ...................... 60 4 8 Logistic regression analysis for likelihood of Fall to Fall retention after one semester of attendance with inclusion of HS GPA ................................ ............. 60 4 9 Logistic regression analysis for Fall to Fall retention with attendance both semesters ................................ ................................ ................................ ........... 61 4 10 Results of Regression Function for First Year College GPA .............................. 61 4 11 Pell status and first year college GPA for students attending Fall 2013 and Spring 2014 ................................ ................................ ................................ ........ 62 4 12 Fall 2013 cohort retention by first year college GPA ................................ .......... 62 4 13 Student retention as a function of credit hours earned versus success of credit accrual ................................ ................................ ................................ ...... 63

PAGE 9

9 4 14 Results of regression function for first year college GPA with inclusion of percentage of accrued credit. ................................ ................................ ............. 63 A 1 Effect of BSC1005 on Fall to Spring retention ................................ .................... 85 A 2 Effect of BSC2010 on Fall to Spring retention ................................ .................... 85 A 3 Effect of MAC1105 on Fall to Spring retention ................................ ................... 85 A 4 Effect of ENC1101 on Fall to Spring retention ................................ .................... 8 6 A 5 Effect of MUL2010 on Fall to Spring retention ................................ .................... 86 A 6 Effect of CHM1030 on Fall to Spring retention ................................ ................... 86 A 7 Effect of CHM1045 on Fall to Spring retention ................................ ................... 87 A 8 Effect of ESC1000 on Fall to Spring retention ................................ .................... 87 A 9 Effect of BSC2093 on Fall to Spring retention ................................ .................... 87 A 10 Retention vs percentage of successful credit accrual ................................ ......... 88

PAGE 10

10 LIST OF FIGURES Figure page 2 1 Relationship of high risk courses and first year drop outs within the existing theoretical frameworks ................................ ................................ ....................... 37 3 1 Gatekeeper/Gateway Courses with High D/F/W Rates ................................ ...... 50 4 1 Total hours earned by college GPA for students attending both semesters ....... 64 4 2 Total hours earned by college GPA for students retained Fall 2014 ................... 65 4 3 Percentage of successful credit hour accrual versus total credit hours earned. ................................ ................................ ................................ ............... 66

PAGE 11

11 LIST OF ABBREVIATIONS AA Associate of Arts degree. Two year degree which is accepted for transfer to a four year institution. AS Associate of Science degree. Terminal degree not intended for transfer to a four year institution. D/F/W Grades of D, F or a Withdrawal from a course First in Series Courses which must be taken first before being allowed to k FTIC First time in college Full time Enrolled in at least 12 college credit hours at the beginning of a semester GPA Grade Point Average Gateway/gatekeeper courses First in series college courses which must be passed to advance in a given major High risk courses Courses which have 20 percent or greater D/F/W rates HS High School Opt out The option to bypass developmental math, English, or reading even if test scores are indicative of need Par t time Enrolled in less than 12 college credit hours at the beginning of a semester

PAGE 12

12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the De gree of Doctor of Education THE RELATIONSHIP OF HIGH RISK COURSES TO FIRST YEAR DROP OUT S AT A SMALL RURAL COMMUNITY COLLEGE By Terolyn Lay December 2017 Chair: Dale F. Campbell Major: Higher Education Administration Student attrition, or drop ping out, from college is a source of concern as it impacts not only the student but the institution as well. Current literature while rich in student retention studies at four year institution s provides little insight into factors impacting student retentio n at rural community colleges. This study, guided by the framework of academic momentum, is intended to fill a gap in rural community college literature. This study evaluated high risk courses, in correlation with selected demographic and academic variable s, for their contributions to first year student drop out risk. Data was collected to as s ess a cohort of first time in college, associate of arts or associate of science degree seeking students attending a small, rural community college located in the panh andle of Florida. Regression analyses were utilized to understand the effect of high risk course taking behavior, in conjunction with other student attributes, on retention rates after one semester and one year of attendance. Additionally, the effect of hi gh risk courses on college GPA as a measure of student success was examined as was the effect of credit accrual. Results showed that high risk courses, apart from Intermediate Algebra, were not a factor in retention rates. High risk courses were a

PAGE 13

13 factor i n student success as measured by college GPA. Successful accrual of credit was found to be the best measure of student success and retention. The results of this study provide a deeper understanding of factors associated with first year drop out risk which can assist in advising and monitoring poor performing students before they exit the institution.

PAGE 14

14 CHAPTER 1 INTRODUCTION In 2009, the Lumina Foundation challenged all higher education institutions to help meet their goal of 60 percent quality after, President Barack Obama presented higher education with the American Graduation Initiative, a plan a imed at community colleges to, in part, increase college graduation rates (White House Office of the Press Secretary, 2009). These challenges, coupled with state generated accountability measures such as performance funding, to retain and graduate students within a reasonable timeframe have colleges and universities working feverishly to increase student retention, persistence and completion rates. High attrition, or drop out, rates pose a serious problem, especially at the community college level. Over sev en million students, pursuing either an Associate of Arts ( AA ) or Associate of Science (AS) degree, enrolled in community colleges in the Fall of 2014 (American Association of Community Colleges [AACC], 2016). Unfortunately, only slightly more than half ( 54.7%) of these students returned to college the following fall (ACT, 2015, p. 3). This attrition rate for community college students has been consistent for nearly the last decade (McIntosh & Rouse, 2009; Schuetz, 2005, 2008). Studies into reasons for l ack of persistence/retention have identified numerous variables which may have an impact on student retention. High school grade point average (GPA) and standardized scores (Habley, Bloom, & Robbins, 2012) have often been used as benchmarks for college su ccess as has first year college GPA (Reason,

PAGE 15

15 2003). Other factors such as first year experience courses (Jamelske, 2009; Stewart, Lim, & Kim, 2015), learning communities, (Pike, Kuh, & McCormick, 2011), academic advising (Darling, 2015), and interaction wi th mentors (Hu & Ma, 2010) have been examined for their impact on retention/persistence. Moreover, demographic factors such as gender, ethnicity and socioeconomic status (SES) (Bailey, Calcagno, Jenkins, Leinbach, & Kienz l 2006; Conway, 2009; Keels, 2013 ; Wolniak, Mayhew, & Engberg, 2012) have also been attributed to student persistence. Most of these studies however, have been focused on four year institutions rather than two year community colleges. Community colleges, as open access institutions, have a vastly different population of students. These institutions are generally commuter (non residential) colleges. Nearly two thirds of their population are part time students with an average age of 28 (AACC, 2016). A study by Crisp and Mina (2012) found t hat more community college students were non white, first generation students requiring remediation, and earning a lower GPA during their first year of college. Additionally, most of the community college experience occurs within the classroom since a majo rity of students rarely remain on campus outside of classroom hours (Barnett, 2011). Many of the students entering community colleges are underprepared academically. According to the What Works in Student Retention (WWISR) survey (Habley, Valiga, McClanah an, & Burkum, 2010), lack of academic preparation for college level work was at the top of the list for causes of attrition. Studies have shown that students with higher high school GPAs, and higher admission test scores are more likely to succeed (Habley, Bloom, & Robbins, 2012) but these students are more likely to attend selective four year institutions.

PAGE 16

16 In the state of Florida, twenty eight institutions serve as the open access gateway to college. Designated as the Florida College System (FCS), these institutions offer both workforce development and baccalaureate degrees, but their primary focus are on AA/AS degree seeking students. Students enrolled in AA programs are most often intending to transfer to a four year program after they either complete t heir AA or have finished the prerequisite courses needed for transfer. In Florida, AS programs are terminal degree programs and are not intended to be used for transfers. In 2015, the FCS student body make up was 35 percent full time, 65 percent part time with 59 percent being female and 58 percent being listed as minority ("FCS Facts," 2016). Colleges within this system range in size from large, multi campus institutions to small, single site, rural institutions. Sixty percent of all community colleges are considered rural (Carnegie Foundation, 2010). These colleges typically have lower enrollments than urban institutions and most often cover a larger geographical area. Unlike their large, urban counterparts, rural institutions face pressures from lower incomes, higher poverty rates and under benchmarks for continued funding (Hicks & Jones, 2 011; Thornton & Friedel, 2016). One such rural institution is the College being utilized for this study. Located in panhandle of Florida, the College serves students from a five county area as well as from the states of Alabama and Georgia. Total student e nrollment for the Fall of 2016 was 2265. Of these, 40 percent were incoming freshmen. The College has a fall to fall retention rate that is consistently in the 55 60 percent range which

PAGE 17

17 exceeds the national average but does not meet the internal standards to which they aspire (M. White, personal communication, October 24, 2016). Identifying the students most likely to drop out can assist the college with increasing its retention rate while ensuring an even greater number of students complete their AA/AS deg rees. Purpose Statement Higher education institutions have, in the last several decades, faced intense scrutiny regarding the number of students that both stay in college (retention/persistence) and finish college (completion). The pressure to retain and g raduate students within a reasonable timeframe has increased as states have used funding to drive institutional behaviors. The reasons for failure to remain in college include lack of student engagement, low GPA, financial pressures, and lack of clear adv ising (Chen and St. John, 2011; Gershenfeld, Hood, & Zhan, 2016; Kuh, Kinzie, Buckley, Bridges, & Hayek, 2006; Kuh, Cruce, Shoup, Kinzie, & Gonyea, 2008; Kuh, 2009) to name a few. However, most of this research has focused on the four year institutions rath er than the community colleges even though the data has consistently shown that close to half of all community college students will not return after their first year (ACT, 2015; McIntosh & Rouse, 2009; Schuetz, 2005, 2008). While many studies have focus ed on GPA, gender, race, or socioeconomic status (Lotkowski, Robbins, & Noeth, 2004; Walpole, 2003) as predictors of student persistence/retention, very few have looked at them in relation to part time/full time status. Moreover, these factors in relation to the number and types of courses being taken have not been studied extensively. Identification of the combination of factors most likely to lead to dropping out would be useful for advising students when enrolling in college, especially during their fir st semester.

PAGE 18

18 The purpose of this quantitative study is to develop a better understanding of the relationship of high risk courses to first year drop out risk for incoming first time in college (FTIC) AA and AS degree seeking students at the College by correlating the number of high risk courses taken with high school GPA, gender, race, Pell status, part time/full time status and first year college GPA. Research Questions This study utilized the foll owing research questions to better understand factors associated with first year drop out risk at a rural community college: 1a. Is there a relationship between the number of high risk courses attempted and first semester student retention, after controlli ng for gender, race, Pell status, and/or full time/part time status? 1b. Is this relationship robust to the inclusion of students' high school preparation as measured via high school GPA? 2a. Is there a relationship between the number of high risk courses attempted and first year student retention, after controlling for gender, race, Pell status, first semester GPA, and/or full time/part time status? 2b. Is this relationship robust to the inclusion of students' high school preparation as measured via high s chool GPA? 3. Is there a relationship between the number of cumulative high risk courses attempted and first year academic success as measured via first year college GPA?

PAGE 19

19 Significance of the Study Most of the research regarding retention and completion h as centered on four year institutions which have different admissions standards and residential housing. Two students. The difference in demographics of the two year student p opulation has been described by many authors (Crisp and Mina, 2012; Kuh, 2006) as has the effect of certain indicators, such as GPA, standardized test scores, race and socioeconomic status. ( Kahn and Nauta, 2001 ; Smith, Droddy, & Guarino, 2011 ). A few stu dies have examined the coursework being attempted, but these have been limited. (Jamelske, 2009; Roska, Davis, Jaggars, Zeidenberg, & Cho, 2009). The emphasis on retention and completion in higher education is not limited to a single school type nor a sin gle state. Thirty two states have some form of performance funding in place (National Conference of State Legislatures website, 2015) with retention and/or completion rates as part of their funding formula. For rural serving institutions, the loss of a sma ll number of students translates to a large percentage drop in retention rates. The re sults of this study can provide rural institutions with a deeper understanding of the relationship between high risk courses and first year drop out risk which can provi de additional ammunition to student advisors when registering and/or counseling students regarding the best pathway to remain in college and achieve their AA or AS degree. Organization of the Study The analyses of course taking behavior as a factor of firs t year drop out risk is organized into five ch apters. Chapter two provides a review of the literature, including current research on student persistence and retention, benchmarks for measuring

PAGE 20

20 success, and the uniqueness of the rural college in higher educ ation. C hapter three describes the research methodology utilized in this stud y, including the setting, data collection, variables, and analytic procedures. Chapter four presents the findings associated with the research questions, as well as additional ob servations which support the framework guiding this study. C hapter five provides a di s cussion of the results as they pertain to course taking behaviors, implications for the institution, recommendations for further research, and limitations of the study. This final chapter highlights the contribution this study offers to the student retention efforts at small rural community colleges

PAGE 21

21 CHAPTER 2 LITERATURE REVIEW The literature review was conducted to provide the background information regarding factors which are thought to contribute to student persistence/retention. This literature review is presented in four sections. Section one addresses the reasons for student attrition or dropping out, from college. Section two offers a review of the liter ature on the factors being addressed in this study cited as affecting student success. Section three describes rural community colleges and their importance to the communities they serve. Lastly, section four describes the framework of academic momentum wh Reasons for Dropping Out Reasons for students dropping out have been studied for over a half a century. relat (Spady, 1970, p. 64). Spady (1970) was the first to examine the interaction of student attributes, such as interests and skills, with the external expectations and influence s of faculty and peers. Spady posited that extrinsic rewards such as grades were more useful for negotiations associated with career opportunities. The development of intellect, however, fell into the category of intrinsic rewards. Here the conditions of normative congruence, having a compatibility with the academic environment, and friendship support, establishing close relationships within the system, were necessary for social integration (Spady, 1970, p. 77). Spady suggested that the chance of droppin college experience and their commitment to the college.

PAGE 22

22 ) and Tinto & Pusser ( 2006) envisioned dropping out as a process built on ch aracteristics of 1) the individual, 2) their interaction with the college setting, and 3) the higher education institution itself. While numerous that ability and personality were two of the most important individual characteristics. largely determine how long and how well they persist. Whether an institution is public or pri vate, two year or four year, or small versus large, plays a role in overall drop out rate. However, the climate of the institution and the support given to its students play a role in th e final analysis, Tinto & Pusser (2006) concluded that the actions of the institution have as much to do with student retention as does the attributes of the students themselves. An early study by Pascarella and Terenzini (1980) found that for freshmen, th e quality of student faculty interaction both in and out of the classroom was as significant of a predictor of attrition as was the quality of their peer relationships. Milem and Berger interactions with faculty influenced their perception of the institution which, in turn, affected their decisions to stay in college. In a meta analysis of the literature, Robbins et al. (2004) nd faculty were highly correlated with student persistence. These studies, however, were conducted at four year institutions where most of the students were enrolled full time and resided on campus. Community college students, on the other hand, possess different attributes than those that begin their

PAGE 23

23 college careers at four year institutions. Crisp and Mina (2012), in an analysis of a Beginning Postsecondary Students Longitudinal Study (BPS: 04/09) follow up, found that community college students in gen eral, were more likely to be: African American or Hispanic; financially independent; first generation college students; less academically prepared; working part or full time during college; having lower degree aspirations; attending college part time; de laying enrollment into college following high school; receiving less financial aid; and earning a lower GPA du ring the first year of college. (p. 154) Unfortunately, students attending a community college are more likely to require some form of remediation before entering a credit bearing course. In examination of thirds or more of community college students enter college with academic skills weak enough in at least one major subject area to threaten Benchmarks for Success According to the AACC (2016), 45 percent of all U.S. high school graduates are enrolled in community colleges. Of these, 41 percent are first time freshmen, over half are women, and less than 50 percent are Caucasian. Unfortunately, the percentage of these students who persist and obtain a degree is staggeringly low. The U.S. Department of Education (USDOE) data, which consists only of first time, full time students, indica tes that 21 percent of these students complete in three years. The National Student Clearinghouse (NSC) paints a somewhat rosier picture in that 57 percent of these same students complete in six years (Juszkiewicz, 2015). However, both the USDOE and the NS C report that part time attendees have a completion rate of less than 25 percent. The inability to retain and complete students is a source of concern for the community colleges.

PAGE 24

24 A multitude of markers have been used in the study of student persistence and retention. Student readiness, student engagement, psychometric measurements, economic variables, and academic performance are some of the indicators studied to better predict the likelihood of student retention and completion at the college level. This s tudy will focus on the following: High school grade point average, gender, race/ethnicity, Pell status, full time/part time status, first semester and first year college grade point average, and gateway/gatekeeper courses. High School Grade Point Average (GPA) Cumulative high school GPA has been linked to first year college persistence (Astin & Oseguera, 2012; Bean & Metzner, 1985; Ishitani, 2006). Ishitani (2006) used national, longitudinal data sets to study the attrition and completion rates of 4,427 s tudents at four year institutions. He found that high school class rank was significantly linked to student persistence. Additionally, Ishitani (2006) found that departure rate was linked to the class rank quintile with more students in the lowest quintile (bottom 25 percent) leaving their second year while students in the third lowest quintile (bottom 50 percent) left more often during the third year of college (p. 876). Jackson & Kurlaender (2014) found that high school GPA was a useful predictor for col lege success over other measures of readiness such as a need for remediation. In a study of 1,140 first time community college students, Feldman (1993) found that high school GPA was the strongest predictor of dropping out. Zajacova et al., (2005) reporte d that students who enrolled for their second year of college had high school GPAs that averaged 0.4 points higher than those who dropped out. A recent report on postsecondary readiness and attainment ( Balfanz, DePaoli, Ingram, Bridgeland, & Fox, 2016 ) u

PAGE 25

25 Longitudinal Study of 2002 (ELS:2002) to examine the correlation between HS GPA and student success. The study found that more than 76 percent of students with a HS GPA of 3.50 or greater (A average) had earned a ba Fifty percent of those with a HS GPA between 3.00 and 3.49 (B average) held a HS GPA), 12.4 percent (2.00 to 2.49 HS GPA), and a mere 3. 3 percent for tho se with a HS GPA less than 2.00 (p. 28). Gender Gender has been studied extensively in conjunction with student retention rates. Early studies by Bean (1980) found that gender was significant in retention at four year institutions. Subs equent studies by Pascarella, Smart and Ethington (1986), Voorhees (1987), and Feldman (1993) all reported that gender was a significant factor in community college student retention with females being retained at higher rates than males. However, later re search has muddled the impact of gender on retention. Bailey et al. (2006) found that in community colleges where women comprised over 50 percent of the student body, women were retained at lower rates than men. Laskey and Hetzel (2011) found no signific ant effect of gender on retention in a study of at risk students entering higher education at a private, four year institution. Likewise, a study of 1,800 college students by DeNicco et al. (2015) found no significant relationship between gender and first year retention at a community college. These latter studies have, however, shown that there is a significant relationship between gender plus ethnicity and college persistence/completion.

PAGE 26

26 Race/Ethnicity The relationship between ethnicity and college atta inment has been shown to be significant. Both Adelman (2006) and Bowen Chingos, and McPherson (2009) noted that higher numbers of whites and Asians entered college when compared to Blacks, Hispanics, and American Indians. An analysis of degrees conferred between 2002 and 2013 by the National Center for Education Statistics (NCES) showed that Blacks and degrees earned. Blacks increased from 12 to 14 percent and 10 to 11 p ercent, respectively. Hispanics had increased from 11 to 16 percent and 7 to 11 percent, respectively. Whites, on the other hand had seen a decrease in their total shares of degrees: 71 to 62 percent and 76 to 69 percent, respectively. (National Center for Education Statistics [NCES], 2016 b para. 4 5) However, the gap between the races continues to be large. Pell Status Low socioeconomic status (SES) is most often correlated with s tudents who come from low income or poverty level backgrounds. Low income is considered 200 percent of the federal poverty level, and poor is defined as 100 percent of the poverty level (National Center for Children in Poverty, 2014) In the 48 contiguous states a family of four making less than $2 4 ,6 00 in 2017 is considered at the federal poverty level, and $4 9 ,2 00 is considered low income (Health and Human Services Department, 2017). To be eligible for Pell grants the total family income must be $50,000 or less for the 2017 2018 school year. However, most Pell money go es to those with a total family income of less than $20,000 (Scholarships.com, 2017).

PAGE 27

27 Research has consistently found a positive association between grant aid and student persistence (Bettinger, 2004; DesJardins & McCall, 2010; Dynarski, 1999). I n studies of low income students, Pell Grant receipt had the greatest impact on continued enrollment (Alon, 2011; U.S. Department of Education, 2011). A study of Louisiana colleges between 2006 to 2009 found that students with Pell Grants were retained at comparable rates to wealthier students (Crockett, Heffron & Schneider, 2011). Additionally, Pell status was found to be the most significant determinant of student persistence of first time full time undergraduates during both the first and second years of enrollment (Khuong, 2014). In 2012, Congress enacted several policy changes for Pell Grant eligibility (HR 3671). Two of these: 1) lowered lifetime maximum number of hours or semesters, and 2) reduction in maximum Estimated Family Contribution, contrib uted to a negative effect on enrollment in three states with high rural populations (Katsinas, Davis, Friedel, Koh, & Grant, 2013). Students in Alabama, Arkansas, and Mississippi who lost their Pell eligibility were faced with a tuition bill they could n ot pay. This resulted in many of them dropping out rather than completing their AA degree (Katsinas, et al., 2013). Full time vs. Part time Status Research has demonstrated that full time status is correlated with higher levels of retention and completion (Adelman, 1999; Bailey, Calcagno, Jenkins, Kienzl, & Leinbach, 2005). Feldman (1993) noted that first time in college students who were enrolled part time were 2.23 times more likely to drop out during their first year at the community college. A study c onducted by Rodriguez (2013) found that the type of first semester enrollment had a significant effect on continued academic success. Part time

PAGE 28

28 students were almost two times less likely to complete their two year degree as compared to their full time coun terparts (Rodriquez, 2013). Unlike the majority of undergraduates at four year institutions, only 40 percent of students attend full time at two year institutions (National Center for Education Statistics [NCES], 2016 a ). These students often have full t ime jobs, a family or both which may negatively impact their college persistence (Fairchild, 2003; Schmid and Abel, 2003) Forman (2009) found that success as a part time student was influenced more by age than by external factors. Her study showed that f or every year of delayed enrollment (up to four years) at a community college, the participants were two and a half times more likely to receive their AA degree than their younger counterparts. Analysis of NCES data, however, showed that only 4.6 percent of first time in college students over the age of twenty five who attend part time obtained an AA degree within three years (Complete College America [CCA], 2011). First semester and First year College GPA First semester and first year college GPA have al so been shown to influence student retention rates (DeNicco, Harrington, & Fogg, 2015; Nora & Crisp, 2012; Zajacova, Lynch, & Espenshade, 2005). The importance of the first year college GPA in student retention has been well documented for 4 year instituti ons (Attewell, Heil, & Reisel, 2011; Hu, McCormick, & Gonyea, 2012; Reason, 2003). DeNicco, Harrington & Fogg (2015) tracked students who first enrolled in a community college in 2006 and followed their college career through 2011. They found that a one po int increase in a year GPA above the average GPA of all freshmen led to a nearly ten percent increase in student retention probability.

PAGE 29

29 The literature on the importance of first semester GPA is less robust but the findings have been sign ificant. Stewart, Lim and Kim (2015) noted that the first semester GPA was highly correlated with persistence. Gershenfeld, Hood and Zhan (2016) found that the first semester GPA was a significant predictor of retention and completion, especially for unde rrepresented students. Students with a first semester GPA of 2.33 or below were at a significant risk for dropping out and were half as likely 4.0 (Gershenfeld, Hood, & Zhan, 2016). Research by Musoba and Krichevskiy (2014) found that first semester GPA was a significant factor in deciding to drop out by Whites, but not by Blacks or Latinos. Gateway/Gatekeeper Courses Research concerning gateway, or gatekeeper, courses and their effect on retention has successfully completed to move forward in a given major. In some cases, gateway major. Amongst the 2016 high school graduates that took the ACT, 61 percent are considered college ready in English, 44 percent in reading, 41 percent in math, 36 percent in science and only 26 percent ready in all four areas. With the exception of science, the other three subject areas have decreased between 5 to 8 percent since 2012 (ACT, Inc., 2016, p. 4). Adelman (2006), in an analysis of the data from the National Center for Education Statistics (NELS:88/2000) study, examined the success of completion of gateway courses by the end of their second year in college. He found that successful participation in these gateway course s directly impacted the odds of completing a

PAGE 30

30 literature. An in teresting study by Musoba and Krichevskiy (2014) found that success in the first math course was significantly associated with continued college enrollment by Blacks and Latinos, whereas English composition was not. Math was more likely to determine drop o ut status whereas English determined graduation status Introductory performance in these courses may have a negative effect on self confidence which may lead to an inability for students to progress through their degree requirements. Unfortunately, students attending a community college are more likely to require some form of remedia tion before entering a credit bearing course. In thirds or more of community college students enter college with academic skills weak enough in at least one major subject area to thre aten their ability to succeed in college level A 2009 Virginia study examined gatekeeper (developmental) course success by first time college students who required some form of remediation in either English, math, reading or some combinat ion of the three subjects. The researchers found that students were more likely to enroll in math gatekeeper courses rather than English or reading. Over a third of those students whose placed in developmental courses failed to enroll in any gatekeeper co urses for the semester. Additionally, over half the students who did enroll in gatekeeper courses failed their courses (Jenkins, Jaggars, & Roksa, 2009). In a Florida study, Calgano, Crosta, Bailey & Jenkins (2007)

PAGE 31

31 found that students who passed their col lege level writing course the first time, after successful remediation, were twice as likely to graduate than those who did not. In 2013, the Florida legislature passed Fla. Stat. 1008 § 30 which restructured 9th grade in a Florida public school in 2003 2004 or thereafter and who earned a standard Florida high school diploma; or students who are serving as active duty Postsecondary Education, 2013). For the Florida state colleges who primarily serve these students, th e opt out provision may increase the risk of high attrition rates in gatekeeper courses such as Intermediate Algebra and English Composition I. The Rural Community College Census B ureau defines rural as that which is left after defining the urban areas (Ratcliffe, Burd, Holder, & Fields, 2016) whereas the U.S. Department of Agriculture considers it to be a combination of open countryside, towns with less than 2,500 people, and urban areas whose populations range from 2,500 to 50,000 but are not part of a metropolitan area (2016, para. 1). In 2005, the Carnegie Foundation for the Advancement of Teaching classified rural serving institutions into three categories: 1) small, less than 2,500 students; 2) medium, 2,500 7,500 students; and 3) large, greater Community College Alliance found that rural community colleges served 37 percent of students enrolled in a two year institution (2010). Rural community colleges face unique problems when compared to larger, urban institutions. Most rural colleges serve small populations within a much larger geographic

PAGE 32

32 area (Pennington, Williams, & Karvonen, 2006). Student s often commute long distances, lack adequate high school preparation and have inconsistent access to technological resources which hampers their ability to persist (Katsinas & Moeck, 2002). Hardy & Katsinas (2007) noted that even rural community colleges differed from each other based on their size. Small rural colleges faced greater obstacles in diverse areas such as providing employment services, child care services, and weekend courses to name a few. Several key challenges to small rural serving instit utions, according to their depressed tax bases, slim margins for error, meeting the needs of diverse student populations and the recruitment and retention of quality employe (2011, p. 31). Nevertheless, all small rural institutions strive to offer both the courses necessary for continued academic success in a four year program and the vocational training at the heart of their workforce programs. This task o often hampered by the lack of adequate funding. The disparity is often linked to the lack s, 2007, p. 15) as well as a lack of understanding of the unique n eeds of rural institutions. residents, state legislators, and policy makers, as catalysts for sustaining high quality of life opportunities for rural America" (Miller & Tuttle, 2007, p. 118). These institutions provide access to educational advancement and cultural enrichment for a population serving community college enjoy s both the challenge and the unique calling to serve its community by providing opportunities to

PAGE 33

33 (Blanchard, Casados, & Sheski, 2009, p. 27) Theoretical/Conceptual Frameworks Historical Frameworks In 1975, Vincent Tinto introduced the interactionalist theory of student departure coupled with their commitment to said institution, gr eatly influence the decision to remain or leave. Tinto (1993) later proposed that the quality of faculty student students were instrumental in retaining students unt il graduation. Tinto developed his theory based on full time, residential student behavior at four year institutions Bean and Metzner (1985) developed a model based on the nontraditional student attending a four year institution According to their mod el, nontraditional students are older than twenty four years, commuters, and/or enrolled part time. Attrition decisions by these students "were based on four sets of variables: (a) academic; (b) background; (c) psychological; and (d) environmental. They em phasized that environmental variables were the most influential in dropout decisions for nontraditional students." (Morrison & Silverman, 2012, p. 74) According to the model, environmental variables such as finances, number of hours worked, and family coul d either be positive or negative influences on remaining in college. Even in the face of positive academic success, negative environmental variables would inevitably lead to the student dropping out of college (Bean & Metzner, 1985). Braxton, Hirschy and McClendon (as cited by Braxton, 2014) developed the

PAGE 34

34 other previously developed models such as that of Bean and Metzner. In this model, student persistence is affected motivation and self efficacy. The interaction of these characteristics with the institutional environment have a direct effect on academic development whereas the external environment has no effect on academic development but does have a direct effect on persistence (Braxton, 2014, p. 111 ). While Braxton Hirschy, and McClendon come closer to mirroring the attributes of community college students in their model, all of these aforementioned researchers used either the four year institution exclusively or utilized both four year and two year institutions in the development of their theories None of these models focus solely on the community college student. What is also lacking in these models is the ef fect of course taking behaviors on student persistence and ultimately completion. Adelman (1999, 2006) examined demographic characteristics, high school performance, and postsecondary performance of a high school cohort from graduation in 1982 through 1993 to determine the likelihood of completion of a four year degree. With this cohort, Adelman (1999) noted and early academic performance Adelman (2006) replicated his study with a cohort that graduated in 1992 and were tracked through 2000. Even though the cohort was from a different decade and had supposedly benefitted from education reform efforts, the results of his second study mirrored those of the first Adelman (2006 ) found that students with low first year college GPAs and low number of college credits failed to retain academic momentum.

PAGE 35

35 Academic Momentum Framework The academic momentum perspective posits that earning college level credits quickly fuels persistence which ultimately ends in completion. Adelman (1999, 2006) high school, the rate of withdrawals/repeats, and the number of college level credits earned per semester. Stude nts who earned twenty or more credit hours within the first year, had few withdrawals/repeats, and remained enrolled without interruption were more likely to graduate than those did not. Of all these factors, however, credit hour accrual was the most impor tant in student retention and completion. Several studies have shown that the completion of coursework during the first showed that earning less than 20 credits in the first y positive correlation between the number of first semester credit bearing hours and transfer rates from the community college to a four yea r institution. In his study, students with 12 or more credits the first semester had higher transfer rates than did those taking a lower number of credit hours. Likewis e, Freeman (2008) noted that first year credit hour attainment, as well as withdrawals/r epeats and continuous enrollment, were accurate measures of academic momentum at a historically Black college. Attewell, Heil, and Reisel (2012) found that students at either a 2 year or 4 year institution who attempted more credit hour courses their fi rst semester of college had the greatest level of academic momentum even without successfully passing said courses. Recently, Attewell and Monaghan (2016) found that students (who worked less than 30 hours per week) who attempted 15 credit hours from the onset, or those

PAGE 36

36 who increased from less than fifteen to fifteen or more during their second semester were more likely to complete a degree within six years which is the current benchmark for student success. In all cases, accumulation of credit hours indi cates positive academic momentum which reduces the likelihood of dropping out. The theories of Tinto, Bean & Metzner, and Braxton, Hirschy, and McClendon include measures of student engagement/integration in their models as well as institutional attributes which may contribute to student retention The framework of academic momentum does not discount the importance of these measures, but focuses on specific measures of academic achievement and student demographics This s tudy takes advantage of a small overlap of the measures often used to examine student retention to explore the relationship of high risk courses to first year drop out risk (Figure 2 1). Summary of Existing Research The lack of student retention/persistenc e in higher education has been studied extensively with many plausible reasons given as to why students drop out of college. The often unique demographics of rural community colleges and the behaviors of their student population has been underrepresented in the literature and provides the basis for the focus of this study. The theoretical concept of academic momentum fits well with risk courses impedes the accumulation of attrition from college.

PAGE 37

37 Figure 2 1 Relationship of high risk courses and first year drop out s with in the existing theoretical frameworks

PAGE 38

38 CHAPTER 3 METHODOLOGY The purpose of this study was to develop a better understanding of the relationship of high risk courses to first year drop out s for incoming first time in college AA and AS degree seeking students at the College by correlating the number of high risk courses attempted with student retention. This chapter revisits the resear ch questions and then describes the setting, data collection, sample population, and the variable utilized in the study. Research Questions This study utilized the following research questions to better understand factors associated with first year drop out risk at a rural community college: 1a. Is there a relationship between the number of high risk courses attempted and first semester student retention, after controlling for gender, race, Pell status, and/or full time/part time status? 1b. Is this relat ionship robust to the inclusion of students' high school preparation as measured via high school GPA? 2a. Is there a relationship between the number of high risk courses attempted and first year student retention, after controlling for gender, race, Pell s tatus, first semester GPA, and/or full time/part time status? 2b. Is this relationship robust to the inclusion of students' high school preparation as measured via high school GPA? 3. Is there a relationship between the number of cumulative high risk cours es attempted and first year academic success as measured via first year college GPA?

PAGE 39

39 Setting The study was conducted with data from a rural, public FCS College (henceforth referred to as the College) located in the Florida Panhandle. The C ollege serves a f ive county 3,382 square mile district, as well as the border counties of Alabama and Georgia. The total population of the five county area is estimated to be 89,250 (U.S. Census Bureau, 2017) The five county population demographics mirror that of the Col lege, with 80 percent being white, 16 percent being black, and the remaining 4 percent being mostly Hispanic. These demographics are not comparable to the state demographics, nor are they representative of all 28 state colleges (Table 3 1). Twenty three percent of the area population lives at or below poverty level. While three quarters of the residents are high school graduates, less than According to the College website, agriculture an d forestry are the major i ndustri es within the five county area. The next largest employment sector is corrections, with four state prisons and a federal correctional facility being found with in the service area. The College, like other state colleges in Florida, is an open admission institution. Of the 28 colleges that comprise the Florida College System (FCS), only six classify as rural serving colleges with these being a mix of the small, medium, and large Carnegie designations (Carnegie Foundation, 200 5). The majority of students who attend the C ollege are natives of this five county district. In contrast with most urban community colleges, 61 percent of the student body are between 17 to 24 years old. The remaining age groups are represented as follows: 11 percent ages 25 to 29, 24 percent ages 30 to 49, and 4 percent ages 50 and above (College website)

PAGE 40

40 The College has a successful TRIO program for first generation college students. The College also has an award winning tutoring center onsite to provide one on one assistance to students as well as group study sessions for selected classes. The College has recently partnered with CareerSource to provide an onsite career resource center which offers ca reer assessments, resume and interview preparation, and even training funds for specific populations of individuals. While the C ollege is ranked highly for completion rates within three years, retention rankings have not been as successful. Overall fall t o fall retention rates for first time in college AA or AS degree seeking students fell ten percentage points between 2011 to 2015 (M. Hughes, personal communication, November, 2015). Understanding the variables which have the greatest impact on retention r ates is important to the C ollege as it works to reverse the trends previously noted. Data Collection Archival data was provided by the Dean of Assessment, Compliance & Grants at the College for all information other than high school GPA. The decision as to which information was collected was a collaborative effort between the researcher and the Dean after studying in house data covering a period of three years which indicated high withdrawal/failure rates for certain classes. The other information, high sch ool GPA, gender, race, Pell status and college GPA, are measures often used when studying retention. Inclusion of full time versus part time enrollment for each semester was necessary since over 50 percent of the students at the College are now classified as part time. Any differences that are due to this status will be of great interest to the College. The data set also included whether the student was seeking an AA degree or an AS degree. AA degree seeking students are used in determination of performan ce

PAGE 41

41 funding levels for the measures of retention and completion. A comparison of their retention levels versus those of AS degree seeking students may provide additional information when comparing classes attempted. Sample Population The sample population for this study were first time in college freshmen entering the College in the Fall of 2013 (n=410). The 2013 cohort was selected for exploring the impact of high risk courses on student attrition because they did not have the choice to to take the Freshmen orientation course were omitted from the data set. Due to the nature of record keeping at the College, this researcher accessed the database to retrieve the high school GPAs of the participants. During the retrieval process it was discovered that five students were coded incorrectly, four were international students with no equivalent HS GPAs on record, and thirty seven held GEDs which also have n o equivalent HS GPAs. These forty six individuals were removed from the data set which left a sample population of n=364. After visually cross checking names and identification numbers to confirm data accrual was correct, all identifying information (name, social security number and/or college ID number) was removed. High Risk Courses Classes included in the data were designated as high risk courses eithe r due to a high D/F/W rate and/or by their first in series designation. The D/F/W rate for these courses was 20 percent or higher These rates are consistently above the average D/F/W rates for all general education courses at the College (Figure 3 1). The courses were:

PAGE 42

42 BSC 1005 Introduction to Biology BSC 2085 Anatomy & Physiology I BSC 2010 Integrat ed Principles of Biology I CHM 1030 Introduction to Chemistry CHM 1045 Chemistry I ENC 1101 English Composition I ESC 1000 Earth Science MAT 1033 Intermediate Algebra MAC 1105 College Algebra MUL 2010 Music Appreciation BSC 1005, ESC 1000, EN C 1101, MAT 1033, MAC 1105 and MUL 2010 are general education courses which have traditionally high failure rates. Successful completion of MAT1033 is required for all students who have taken developmental math courses before being admitted to MAC 1105. CH M 1030 is a precursor to CHM 1045. For students lacking high school chemistry, CHM 1030 must be passed to enter CHM 1045. BSC 2085 is high risk for the students seeking the AS Nursing degree. BSC 2085 is also a first in series course as are BSC 2010 and C HM 1045 for STEM students to progress in their chosen degree fields. Measures Dependent Variables The dependent variables (Table 3 2 ) are fall to spring retention and fall to fall retention for research questions one and two. The variable was operational ized to reflect either as enrolled in Spring of 2014 or not enrolled for the first model. The variable was also operationalized to reflect as enrolled in Fall of 2014 or not enrolled for

PAGE 43

43 the second model. Enrollment status was included in the data set prov ided by the College. The dependent variable for research question three was first year college GPA. The variable was operationalized to reflect GPA on a scale of 0.00 to 4.00. Independent Variables The number of high risk courses attempted during the Fall semester served as the independent variable (Table 3 3 ) for research question one. The number of high risk courses attempted in the first year served as the independent variable for research question two. For research question three the total number of hig h risk courses served as the independent variable. For research questions one and two, the independent variables were operationalized as enrolled in none, one, or two plus high risk courses. For research question three, these were operationalized as no ne, one, two, or three plus high risk courses. Covariates The inclusion of high school GPA, gender, race, Pell status, and part time/full time status (Table 3 4 ) allowed for examining the bivariate relationship between each characteristic and course taki ng behavior. Additionally, descriptive statistics of each characteristic provided a complete picture as to the population being examined. Detailed reasons for including these characteristics were provided in Chapter 2, but a brief overview is presented he re. High school GPA has been shown to be a strong predictor of college persistence (Hoffman & Lowitizi, 2005; Stewart, Lim, & Kim, 2015) as well as being a reliable

PAGE 44

44 measure of academic achievement in college (Habley et al., 2012). Inclusion of high school GPA in the analysis increased the reliability and validity of the instrument. The effect of gender and race on persistence in higher education has been ambiguous. Some researchers have found that both gender and race are linked to college persistence (Keel s, 2013;) while others have found gender but not race to be significant (Windham, Rehfuss, Williams, Pugh, & Tincher Ladner, 2014). Analysis of these characteristics determined their usefulness in predicting student drop out risk at the College. The effe ct of full time enrollment on persistence has been shown to be positive (Bailey, Calcagno, Jenkins, Kienzl, & Leinbach, 2005; Brooks Leonard, 1991). In comparison, part time enrollment has been found to negatively impact student retention (Schmid and Abell 2003) Since the College now has more part time students than full time students, analysis of these two characteristics was useful in identifying trends within the college population. Analysis Methods Descriptive statistics were utilized to understand the characteristics of the data as well as to locate any missing values. No missing values, other than the aforementioned HS GPAs, were encountered. Due to the low overall percentage of minorities in general, and very low percentage of individual minority gr oups, all minorities were analyzed as a single entity. The software package, STATA, was used for all analytical methods. French, Immekus, & Yen, 2013, p. 145), was used for research questions one and two to examine the relationship between retention and high risk courses. The categorical (dependent) variable for this

PAGE 45

45 study was retention. Logistic regression was the suitable choice for this study since the dependent variable was dichotomous with only two outcomes possible: retained or not retained. Additionally, logistic regression examines the influence of additional variables on the outcomes by estimating the probability of assumptions that must be met for logistic regression include: Independence of errors no duplicate responses between sample groups exist. Linearity in the logit for continuous variables continuous variables and their o utcomes are linear. Absence of multicollinearity no redundancy amongst the independent variables. Lack of strongly influential outliers the predicted outcome should be close to the actual outcome. (Stoltzfus, 2011, p. 1101) O dds ratios (i.e., the chan ge in the odds of an event occurring versus not occurring given a one unit change in an independent variable) were used for ease of interpretation of the logistic regression models (STATA14, 2015). Cohort level r obust standard errors were utilized to addre ss the potential problem of errors that are not independent and identically distributed (Williams, 2015). For research question three linear regression was utilized. Linear regression, es for a examine the relationship between first year college GPA and total number of high risk classes attempted, linear regression was the suitable choice for the ana lysis. Several assumptions must be made when using linear regression if the results are to be generalized for a larger population of students other than those at the College. These assumptions, as laid out by Cohen (2013) are: Independence: The residuals are statistically independent of each other.

PAGE 46

46 Linearity: The relationship between variable can be graphically represented by a straight line. Normal distribution: Variables must have equal distribution in the population. Homoscedasticity: The residuals have constant variance. The Breusch Pagan/Cook Weisberg test, which tests the null hypothesis that the variance of the residuals is homogenous wa s utilized to check for heteroskedasticity of the first year college GPAs I f the p value is very small the null hypothesis must be rejected and the alternative hypothesis that the variance is not homogenous must be accepted (IDRE, 2017). Variables for the test were the fitted values of the cumulative college GPA. Results of the test were chi2(1) =40.98 and the p value was 0.000 therefore the null hypothesis was rejected. The variance inflation factor (VIF) was used to check for multicollinearity The VIF estimates how much the variance of a coefficient predictors mean VIF for all variables was 1.56 which indicates a lack of multicollinearity. These results provide assurance that linear regression is the appropriate model for research question three. Additional Considerations. An an alysis of the ratio of high risk course hours versus total hours enrolled (risk ratio) was run to determine whether this w ould have an effect on the model being studied. The analysis showed that greater than 50 percent of the students were carrying high ri sk course hours at less than 25 percent of their entire load for both the fall semester and for the entire year (Tables 3 5 and 3 6 ) Inclusion of the risk ratio in the logistic regression models did not effect changes in the models therefore the risk rat io was determined to be unnecessary for inclusion.

PAGE 47

47 An analysis of accrued credit was also performed to test the validity of the academic momentum framework guiding this study. Accrued credit was operationalized as a percentage of the total hours attempted versus the total hours earned for the first year of attendance. Inclusion of this variable into the model for research question three had a significant effect on the model and thus was added into the additional findings. Limitations of the Study The prima ry limitation of this study was the small sample size of the cohort used for this analysis. A larger sample size might yield a better picture of the effect high risk courses have on student retention. Applying the model to larger college populations could alter which demographics are most salient in determining risk. of cohorts who could utilize the opt out provision may paint a very different picture as compared to the cohort first time in college freshmen may provide a more robust model. Summary of Methodology This chapter provides an overview of the statistical methodologies to be utilized to address the propo sed research questions. The purpose of this study was to develop a better understanding of the relationship of high risk courses to first year drop out risk This chapter also addresses the data source, the data collection methods, and the variables of int erest for this study. Finally, the chapter concludes with potential limitations that may result in this study.

PAGE 48

48 Table 3 1. Ethnic representation of Florida and its state colleges Location Ethnicity White Black Hispanic Other State College #1 69.2% 10.8% 11.5% 8.6% State College #2 20.3% 34.1% 35.5% 10.1% State College #3 59.9% 12.7% 14.6% 12.9% State College #4 77.8% 14.8% 3.8% 3.7% State College #5 65.4% 12.6% 14.7% 7.3% State College #6 51.8% 11.4% 28.1% 8.6% State College #7 49.3% 23.6% 7.6% 19.6% State College #8 71.1% 7.7% 0.0% 21.2% State College #9 71.0% 11.8% 5.8% 11.4% State College #10 35.3% 17.2% 26.6% 20.9% State College #11 56.6% 14.6% 20.0% 8.9% State College #12 82.1% 11.5% 4.9% 1.6% State College #13 66.1% 9.6% 8.0% 16.4% State College #14 66.9% 8.4% 14.6% 10.1% State College #15 6.8% 15.6% 72.9% 4.7% State College #16 75.2% 17.7% 6.2% 0.9% State College #17 72.0% 7.4% 8.8% 11.8% State College #18 37.0% 24.4% 29.0% 9.6% State College #19 68.2% 4.4% 19.1% 8.3% State College #20 70.3% 13.5% 5.6% 10.6% State College #21 53.9% 15.8% 20.7% 9.7% State College #22 73.8% 7.6% 8.9% 9.7% State College #23 64.9% 12.5% 12.0% 10.5% State College #24 58.2% 13.5% 19.3% 9.0% State College #25 50.4% 15.4% 26.0% 8.1% State College #26 52.4% 9.3% 32.8% 5.5% State College #27 53.1% 28.0% 12.4% 6.5% State College #28 29.5% 16.7% 33.7% 20.1% State of Florida 54.9% 16.8% 24.9% 5.6% Note: Percentages for state colleges represent the most recently available data (2014 2015) as reported by the Florida College System. Percentages for the state of Florida are 2016 estimates an, Neither Hispanic nor Latino, 2 or more races, Pacific Islander, American Indian, or Not Reported. Table 3 2 Summary table of dependent variables Dependent Variables Variable Type Value Fall to Spring Retention Dichotomous 0 = Not Retained ; 1 = Retained Fall to Fall Retention Dichotomous 0 = Not Retained ; 1 = Retained First semester College GPA Continuous 0.00 4.00 First year College GPA Continuous 0.00 4.00

PAGE 49

49 Table 3 3 Summary table of independent variables Ind ependent Variables Categories Variable Type Value Number of High risk Courses Attempted per Semester Zero One Two or More Dichotomous 0 = No; 1 = Yes Number of Total High risk Courses Attempted Zero One Two Three or More Dichotomous 0 = No; 1 = Yes Table 3 4 Summary table of covariates Covariates Variable Type Value Gender Dichotomous 0 = Male; 1 = Female Ethnicity Dichotomous 0 = Minorities; 1 = White Pell Recipient Dichotomous 0 = No; 1 = Yes Attendance Dichotomous 0 = Part time; 1 = Full time High School (HS) GPA Continuous 0.00 4.00 Table 3 5 Risk ratio for st udents enrolled during Fall 2013 Note: N=364. Table 3 6 Cumulative risk ratio for students enrolled both Fall 2013 and Spring 2014 Note: N=300. Risk Ratio Number of Students Percentage 0.00% 0.01% 24.99% 25.00% 49.99% 50.00% 100% 132 110 93 29 36.26% 30.22% 25.55% 7.97% Risk Ratio Number of Students Percentage 0.00% 0.01% 24.99% 25.00% 49.99% 50.00% 100% 42 154 92 12 14.00% 51.33% 30.67% 4.00%

PAGE 50

50 Figure 3 1. Gatekeeper/Gateway Courses with High D/F/W Rates

PAGE 51

51 CHAPTER 4 RESULTS This study evaluates the relationship between high risk courses and drop out risk during the first year of college attendance. It also examines the significance of high school GPA and the first year college GPA as a measure of retention and student success This chapter begins with the descriptive statistics of the student population, followe d by the regression analyses for fall to spring retention (RQ1) and fall to fall retention (RQ2) Next, the analyses of first year college GPA and high risk courses are presented (RQ3). Additional findings as to the effect of credit accrual on academic success and retention are presented followed by summary of the results to conclude the chapter. Descriptive Statistics : Gender, Ethnicity, Enr ollment Status, Pell Status, HS GPA and High Risk Course Enrollment Gender. The Fall 2013 first time in college cohort ( n =364) had a greater percentage of females to males (Table 4 1). The study sample mirrors the sample populations described in the litera ture regarding community college students. Ethnicity. The ethnic make up of the sample population is reflective of the five county area served by the college. The majority of the Fall 2013 cohort were white (74.18%). The remaining 39.29 percent were pred ominantly black (16.21%), with Hispanic (5.22), Mixed race (2.74%), Indian (0.82%) and non Hispanic (0.82%) also represented Table 4 1). This ethnic breakdown is unlike that of the literature, which has shown the community college population to have a grea ter percentage of minority students as compare to white students. Enrollment Status. The majority of the Fall 2013 cohort were enrolled full time (81.87%) as compared to part time (18.13%) (Table 4 1). This pattern of enrollment is

PAGE 52

52 an antithesis of the l iterature in which part time enrollment is the rule, rather than the exception, in community colleges. Pell Status. Over half of the sample population (57.14%) were Pell recipients (table 4 1). The literature suggests that financial need is high in comm unity colleges therefore finding a majority of the Fall 2013 cohort were recipients of Pell grants was not unexpected. HS GPA. High school grade point average (HS GPA) has been used extensively in determining potential college success. The Fall 2013 cohor t had a range of HS GPAs From 1.46 to 4.00 (Table 4 1). The average HS GPA was 3.21 and the majority of the sample population (67.30%) earned HS GPAs between a 3.00 3.99. High Risk Course Enrollment As seen in Table 4 2, over one third of the stud ents were not enrolled in any high risk course during their first semester of college. An equal number of students (36.26%) were enrolled in one high risk course. Slightly more than a quarter of the students (27.47%) were enrolled in two or more of these c ourses. By the end of the first year in college, the majority of students (81.59%) had taken a total of one, two, or three plus high risk courses with the majority taking one high risk course (Table 4 3). Research Question 1: Effect of High Risk Courses an d other Factors on Fall to Spring Retention This section examines the factors which had the greatest likelihood of affecting student retention after one semester of enrollment. An examination of the influence of each high risk course (Tables 4 4; A 3 to A 11) on student retention revealed no individual course had a significant relationship with retention from Fall to Spring apart from Intermediate Algebra (MAT1033). As seen in Table 4 4, students enrolled in

PAGE 53

53 MAT1033 during the fall semester, when controlling for all other factors, were less likely The number of high risk courses taken during the Fall semester were not significant predictors of retention (Table 4 5). Only par t time attendance was shown to 4 6), shows the two significant predictors of student retention from fall to spring are part 1). First time in college students attending part time the first semester are much less likely to return the following spring semester. For each unit increase in HS GPA, students are two times more likely to remain in college beyond the first semester. Re search Question 2: Effect of High Risk Courses and Other Factors on F all to Fall Retention Fall to Fall Retention: Entering Cohort Analysis of the 2013 cohort for potential return in the Fall of 2014 after finishing their first semester of college yields s imilar results to that for Fall to Spring retention with one notable exception. Part time attendance was found to be significant indicator of 7). The addition of HS GPA (Table 4 8) resulted in part time attendance no longer having a significant returning after the first year for the entire 2013 cohort. In both models, the first semester college GPA was th to fall student retention aft er completion of one semester.

PAGE 54

54 Fall to Fall Retention: Students with Continuous Enrollment To examine the relationship of first year course taking behavior on fall to fall retention, only students who attended both fall and spring semesters were included in the analysis. As seen in Table 4 9, taking even one high likelihood of retention v ersus that of a student who did not take a high risk course. A 0.01) had an even greater positive effect on student retention. Two or more high risk courses quadr uples the odds of retention while three or more high risk courses yields a quintupling of the odds. Furthermore, the first year college GPA becomes the most .001 ) with each unit increase in college GPA nearly triplin g the likelihood of retention. For students who attended both semesters, HS GPA is no longer a significant predictor of retention. Research Question 3: First Year College GPA and High Risk Courses as a Measure of Academic Success This section examines the relationship of high risk courses on academic success as measured by the first year college GPA. The regression function for first year college first year had a sign ificant impact on academic success (Table 4 10). The negative coefficient indicates that poorer performance in these classes results in a lower overall GPA. Both HS GPA and Pell status were highly significant ( p relationship with stude nt achievement. However, HS GPA had a positive relationship whereas Pell status resulted in a negative relationship Table 4 11 provides a more detailed picture of the interaction between Pell and first year college GPA. Thirty six percent of the Pell r ecipients who attended both

PAGE 55

55 recipients. More than half the Pell recipients fell below a GPA of 3.00 whereas over half the non recipients were at or above a 3.00 GPA. Ethnicity and part time attendance were less of first year success. Table 4 12 demonstrates the relationship of first year college GPA on first year retention. Of the 364 students who entered in Fall of 2013, 39.84 percent did not return the following year. Of the students who did not return over half (52.41%) had first year college GPA scores below a 2.50 and 64.13 percent had scores below a 3.00. Over half (54.79%) of the returning students had GPAs at or above 3.00 and a total of 82.19 pe rcent of all returning students had GPAs at or above a 2.50. Additional Findings To fully explore the concept of academic momentum, an examination of the number of credit hours earned in relation to retention as well as the relationship between earned hour s, first year GPA and student retention was needed. Additionally, the percent of total credit earned in relationship to both first year GPA and high risk courses could provide a clearer picture of the interaction of these variables. Figures 4 1 and 4 2 s how the effect of college GPA on retention and the relationship of college GPA to accumulated credit hours. For students who attended in both Fall 2013 and Spring 2014 semesters (Figure 4 1), those that took a total of 15 hours or more between the two seme sters have GPAs that are greater than 2.00 with few exceptions. Students with less than 15 hours total have a wide range of GPAs and as the total hours decrease, the range of GPAs increases. When compared with Figure 4 2, one can observe the lack of return ing students with GPAs below 2.00. Additionally, more returning students have a total number of credit hours at 15 hours or above.

PAGE 56

56 An examination of course taking success yielded unexpected effects. Measured in terms of percentage of accrued credit, i.e. how many hours earned versus enrolled (or what was the percentage of course taking success per student), the percentages ranged from zero to one hundred percent accrual. A total of 165 students accrued 20 or more credit hours between the two semesters with 34 credit hours being the maximum earned. One hundred twenty five of these students were 100 percent successful in earning the credit hours they originally attempted and only 14 of these students accrued less than 20 hours. The remaining 54 students with 20 or more hours exhibited a wide range of successful accrual with a high of 96 percent to a low of 60 percent (Table A 12) Figure 4 3 shows the ratio of earned credit hours as compared to the percentage of successful credit hour accrual. The likelihoo d of student retention is much greater for those students who earn a higher number of credit hours and attain a higher percentage of successful credit hour accrual. As seen in Table 4 13, students with greater than 20 hours of earned credit and greater tha n 75 percent credit hour accrual success were more likely to be retain ed. Of the 212 students who returned in the Fall of 2014 64.6 percent had credit hours totaling more than 20 hours. The addition of percentage of accrued credit to the regression analy sis of first year college GPA to determine its relationship to retention (Table 4 13) further solidifies the importance of successfully earning credit in a timely manner. Percentage of accrued credit ( p ship with GPA whereas enrolling in three or more total high risk courses ( p

PAGE 57

57 also correlated to a significant negative relationship wi th GPA. These results indicate that students taking high risk courses are less likely to successfully accrue credit and are more likely to earn lower college GPAs. Summary Of Results The data analyses presented in this chapter are designed to evaluate hig h risk course taking behavior and their impact on student retention of first time in college students in correlation with other significant attributes such as gender, ethnicity, and enrollment status. Additionally, the impact of high school and college lev el grade point averages on student retention are examined. The results from the regression models indicated that neither individual high risk courses or combinations of those were solely responsible for student retention beyond the first semester. Part t ime status was shown to have a negative correlation with retention in all instances. Both HS GPA and the first semester college GPA were shown to have a positive correlation with retention. The total number of high risk courses taken, in conjunction with first year college GPA, demonstrated a negative correlation for retention beyond the first year. The successful accrual of credit hours was also shown to have a positive impact on student retention. These findings, their implications, and recommendations f or future studies will be discussed in the following chapter.

PAGE 58

58 Table 4 1. Descriptive statistics: Fall 2013 cohort Covariates Subdivisions Frequency Percentage Gender Female Male 221 143 60.71 39.29 Ethnicity White Minority Black Hispanic Mixed Indian Non Hispanic 270 94 59 19 10 3 3 74.18 25.82 16.21 5.22 2.74 0.82 0.82 Enrollment status Full time Part time 298 86 81.87 18.13 Pell recipient Yes No 208 156 57.14 42.86 HS GPA 1.46 1.99 2.00 2.49 2.50 2.99 3.00 3.49 3.50 3.99 4.00 7 29 70 134 111 13 1.92 7.97 19.23 36.81 30.49 3.57 Note: N=364. Table 4 2. First semester high risk course taking behavior Note: N=364 Table 4 3. First year high risk course taking behavior Note: N=364 High Risk Courses Attempted Number of Students Percentage Zero 132 36.26% One 132 36.26% Two or more 100 27.47% Total High Risk Courses Attempted Number of Students Percentage Zero 67 18.41% One 108 29.67% Two 98 26.92% Three or more 91 25.00%

PAGE 59

59 Table 4 4. Effect of MAT1033 on Fall to Spring retention Variables Odds Ratio Robust Standard Error z Sig. Female 0.723 0..228 1.03 0.304 White 0.588 0..215 1.45 0.147 Part time 0.259 0.088 3.97 0.000*** Pell recipient 1.143 0.354 0.43 0.667 MAT1033 0.260 0.125 2.80 0.005** HS GPA 2.018 0.575 2.46 0.014* Constant 1.414 1.400 0.35 0.727 Note: N=364. *** p Table 4 5. Logistic regression analysis for Fall to Spring r etention Variables Odds Ratio Robust Standard Error z Sig. Female 0.815 0.253 0.66 0.511 White 0.716 0.245 0.97 0.330 Part time 0.211 0.066 4.98 0.000*** Pell recipient 0.974 0.287 0.09 0.930 One high risk course 0.977 0.324 0.07 0.943 Two or more high risk courses 0.946 0.351 0.15 0.881 Constant 10.682 4.898 5.17 0.000*** Note: N=364. *** p Table 4 6. Logistic regression analysis for Fall to Spring r etention with i nclusion of HS GPA Variables Odds Ratio Robust Standard Error z Sig. Female 0.740 0.232 0.96 0.336 White 0.650 0.229 1.22 0.221 Part time 0.265 0.091 3.88 0.000*** Pell recipient 1.084 0.331 0.26 0.793 One high risk course 0.901 0.305 0.31 0.759 Two or more high risk courses 0.843 0.314 0.46 0.646 HS GPA 2.057 0.570 2.60 0.009** Constant 1.194 1.166 0.18 0.856 *** p

PAGE 60

60 Table 4 7. Logistic regression analysis for likelihood of Fall to Fall retention after one semester of attendance Variables Odds Ratio Robust Standard Error z Sig. Female 1.392 0.324 1.42 0.155 White 1.335 0.351 1.10 0.271 Part time 0.519 0.158 2.15 0.031* Pell recipient 0.717 0.171 1.40 0.162 One high risk course 0.877 0.241 0.48 0.633 Two or more high risk courses 1.440 0.430 1.22 0.222 First semester GPA 1.695 0.196 4.56 0.000*** Constant 0.288 0.142 2.52 0.012* *** p Table 4 8. Logistic regression analysis for likelihood of Fall to Fall retention after one semester of attendance with inclusion of HS GPA Variables Odds Ratio Robust Standard Error z Sig. Female 1.287 0.304 1.07 0.286 White 1.206 0.325 0.70 0.486 Part time 0.654 0.203 1.37 0.172 Pell recipient 0.784 0.189 1.01 0.314 One high risk course 0.827 0.231 0.68 0.497 Two or more high risk courses 1.317 0.397 0.91 0.361 First semester GPA 1.624 0.193 4.07 0.000*** HS GPA 2.014 0.467 3.02 0.003** Constant 0.038 0.032 3.86 0.000*** *** p

PAGE 61

61 Table 4 9. Logistic regression analysis for Fall to Fall retention with attendance both semesters Variables Odds Ratio Robust Standard Error z Sig. Female 1.551 0.451 1.51 0.131 White 1.221 0.372 0.66 0.511 Part time 1.546 0.732 0.92 0.358 Pell recipient 0.771 0.243 0.82 0.411 One total high risk course 3.400 1.693 2.46 0.014* Two total high risk courses 4.382 2.098 3.09 0.002** Three or more total high risk courses 5.048 2.413 3.39 0.001** HS GPA 1.470 0.422 1.34 0.179 First year college GPA 2.915 0.626 4.98 0.000*** Constant 0.008 0.010 4.23 0.000*** Note: *** p Table 4 10. Results of Regression Function for First Year College GPA Variables Coef. Robust Standard Error t Sig. Female 0.124 0.088 1.41 0.159 White 0.203 0.100 2.03 0.044* Part time 0.272 0.137 1.99 0.048* Pell recipient 0.301 0.081 3.72 0.000*** One total high risk course 0.128 0.151 0.85 0.397 Two or more total high risk courses 0.293 0.166 1.77 0.078 Three or more total high risk courses 0.493 0.161 3.07 0.002** HS GPA 0.594 0.107 5.56 0.000*** Constant 1.071 0.424 2.53 0.012* Note: N=300. F (8,291) = 14.77. Prob>F=0.000***. R .01, *** p

PAGE 62

62 T able 4 1 1 Pell status and first year college GPA for students attending Fall 2013 and Spring 2014 First year cumulative college GPA range Did not receive Pell funding Did receive Pell funding 0.00 1.99 5 (3.94%) 36 (20.81%) 2.00 2.49 14(11.02%) 27 (15.61%) 2.50 2.99 30 (23.62%) 42 (24.28%) 3.00 3.49 41 (32.28%) 41 (23.70%) 3.50 3.99 27(21.26%) 18 (10.40%) 4.00 10 (7.87%) 9 (5.20%) Total # of students 127 (100%) 173 (100%) Note: N=300. Table 4 1 2 Fall 2013 cohort retention by first year college GPA Fall 2013 to Fall 2014 Student Retention First year cumulative college GPA range Did Not Return Did Return Total 0.00 1.99 52 (35.86%) 10 (4.57%) 62 (17.03%) 2.00 2.49 24 (16.55%) 29 (13.24%) 53 (14.56%) 2.50 2.99 17 (11.72%) 60 (27.40%) 77 (21.15%) 3.00 3.49 26 (17.93%) 68 (31.05%) 94 (25.82%) 3.50 3.99 13 (8.97%) 35 (15.98%) 48 (13.19%) 4.00 13 (8.97%) 17 (7.76%) 30 (8.24%) Total # of students 145 (100%) 219 (100%) 364 (100%)

PAGE 63

63 Table 4 13. Student retention as a function of credit hours earned versus success of credit accrual Total Credit Hours Earned <20 hours Percentage of Credit Earned vs Attempted Number of Students Retained Number of Students Not Retained Number of Students Retained Number of Students Not Retained 100% 92 (43.39%) 19 (21.59%) 11 (5.19%) 3 (3.41%) 75 99% 41 (19.34%) 7 (7.96%) 8 (3.77 %) 5 (5.68 %) 50 74% 4 (1.89%) 2 (2.27%) 39 (18.40 %) 13 (14.77 %) 25 49% 0 (0.00%) 0 (0.00%) 14 (6.60 %) 20 (22.73 %) 0 24% 0 (0.00%) 0 (0.00%) 3 (1.42 %) 19 (21.59 %) Total # of students 137 (64.62%) 28 (31.82%) 75 (35.38%) 60 (68.18 %) Note: N (Retained)=212; N (Not Retained) = 88. Students must have attended two consecutive semesters to earn 20 or more hours of credit. Table 4 14. Results of regression function for first year college GPA with inclusion of percentage of accrued credit. Variables Coef. Robust Standard Error t Sig. Female 0.110 0.074 1.49 0.138 White 0.064 0.094 0.68 0.494 Part time 0.096 0.128 0.75 0.455 Pell recipient 0.121 0.069 1.76 0.080 One total high risk course 0.267 0.148 1.81 0.072 Two total high risk courses 0.440 0.155 2.84 0.005** Three or more total high risk courses 0.618 0.150 4.12 0.000*** Percentage of accrued credit 0.017 0.002 7.13 0.000*** HS GPA 0.135 0.097 1.39 0.167 Constant 1.414 0.342 4.13 0.000*** Note: N=300. F (9,290) =27.29. Prob>F=0.000***. R squared=0.489. *** p

PAGE 64

64 F igure 4 1. Total hours earned by college GPA for students attending both semesters

PAGE 65

65 Figure 4 2 Total hours earned by college GPA for students retaine d Fall 2014

PAGE 66

66 Figure 4 3. Percentage of successful credit hour accrual versus total credit hours earned.

PAGE 67

67 CHAPTER 5 DISCUSSION AND CONCLUSION In this chapter the purpose of the study and the analytical results which address the research questions are reviewed. Next, the findings regarding characteristics most like ly to predict drop out risk of first time in college (FTIC) community college students, as framed by the current research literature, are discussed. The next section discusses i mplications for institutional practice followed by recommendations for future research. Finally, the limitat ions of the study are addressed with concluding remarks ending this chapter. Purpose of Study Reviewed Rural community colleges, like most other hig her education institutions in the United States, are under pressure to increase retention rates. Numerous reasons, such as lack of on n feld et al., 2016), and unmitigated financial pressures (Kuh et al., 2006), have been identified as causes for the lack of persistence in the community college setting. Gatekeeper/gateway courses (Dougherty et al., 2009; Gainen, 1995; Leinbach & Jenkins, 2008) have been identified as potential stumbling blocks for continu ed success in chosen academic pathways. These courses can inhibit continuous accrual of credit hours which has been implicated in affecting student retention (Adelman, 1999, 2006; Goldrick Rab, 2007). However, observations regarding how the number of these high risk courses taken at one time and their effect on first year retention has not been examined. The purpose of this quantitative study was to develop a better understanding of the relationship of high risk courses to first year drop out risk for incom ing FTIC AA and AS degree seeking students. This study sought to answer three questions. The first was to what

PAGE 68

68 extent does taking high risk courses in the first semester affect student retention. The second question examined the effect of high risk course s on continued retention after one year of college attendance. The final question sought to determine the relationship between high risk courses and first year academic success as measured by college GPA Descriptive statistics and regression analyses were utilized for data analysis. Findings are discussed in the following section. Discussion of Results Course taking behavior and retention. The analysis of the relationship between high risk courses and the likelihood of student retention after the firs t semester provided unexpected results. Individual high risk courses, except for Intermediate Algebra ( MAT 1033 ) showed no significant relationship with student retention. MAT 1033 is both an introductory mathematics course and a gateway course. It has be en shown that gateway courses such as MAT 1033 can be a barrier to student success (Gainen, 1995; Musoba & Krichevskiy, 2014; Offenstein, Moore, & Shulock, 2010). For underprepared students, there is an even greater likelihood of failure in gateway course s. Adelman (1999) showed that the most likely courses to be either failed, repeated, or withdrawn from were mathematics courses. Inability to progress through these courses impedes academic momentum and potentially sets the stage for student attrition. Wh ile MAT 1033 does not satisfy any general education requirement for graduation, it is required for students who do not meet the standards for entry into College Algebra (MAC 1105) and may serve as an elective as well Students who take developmental math m ust successfully pass MAT 1033 before entering MAC 1105. Studies have shown that of students who start in a developmental or remedial math

PAGE 69

69 class only 31 percent will complete the required math series (Bailey, 2009). Additionally, the more developmental cl asses required, the more likely the student is to drop out of college. By the time a student reaches MAT 1033, they may have spent a year in developmental classes, and will have not gained any college level credit towards the math requirements for graduati on Many of these students will experience frustration and burn out which impacts their continued college success. The impact of MAT 1033 on student retention for the Fall 2013 cohort carries even greater weight considering the developmental education legislation enacted in Florida in 2013. As previously mentioned, students who graduated from a public Florida high school, after 2008, or active duty military personnel, have the option to skip developmental education courses e ven if their entrance scores or high school transcripts indicate otherwise. MAT 1033 is the course in which all opt out math students are enrolled. An analysis of the FCS FTIC 2014 cohort who chose to opt out of developmental math revealed a 12.5 percent d rop in passing rates for MAT 1033 (Park et al., 2016). Institutions facing decreasing success in a gateway course are also facing a potential reduction in performance funding monies if these increased failures also decrease the overall retention rate. En rollment in one or more high risk courses during each semester had no significant relationship with overall retention. However, the aggregated effect of taking a total of one or more high risk courses did have a significant impact on student retention. Thi s suggests that students taking high risk courses and passing those courses are attaining academic momentum. As discussed in Chapter 2, academic momentum (Adelman, 1999) emphasizes the timely accrual of college credit, especially during the

PAGE 70

70 first year of a ttendance. Adelman (1999, 2006) found, regardless of incoming student academic achievement, socio economic status, and other student related covariates, that attaining 20 credit hours or more during the first year increased the likelihood of graduation. Do yle (2011) specifically examined credit accrual at the community college level and found a linear relationship between credits earned and probability of transfer to a four year inst itution. Adelman (2006) noted that for community college students, a 20 per cent failure or withdrawal rate reduced the transfer probability by 39 percent. Additionally, Moore and Shulock (2009) found that as the percentage of dropped courses increased, the likelihood of continued enrollment decreased, even when controlling for al l other factors like part time attendance and ethnicity. Significant Characteristics of Drop Out Risk Part time vs Full time Enrollment Part time enrollment is a significant roadblock to retention, even when the overall percentage of part time students is low. The College reports a greater than 50 percent part time attendance rate annually yet the FTIC incoming cohort analyzed for this study was less than 20 percent part time. The disparity in these numbers indicates a need for further evaluation of this f actor. Nevertheless, part time enrollment, especially in the first semester of college, presents a potentially serious impediment to student retention. The absence of part time enrollment as a significant indicator of retention beyond the first semester is most likely attributable to the low percentage of incoming students in this category. Forty one percent of the part time students failed to return after one semester leaving a very small number of part time students remaining within the cohort. The attrit ion rate of the part time students at the College mirrors that cited in the literature for community college students in general.

PAGE 71

71 Full time attendance would be preferable for all college students. In Florida, where performance based funding (PBF) is in ef fect, retention and completion rates are only measured for full time students. Not only is student retention necessary, but the time to completion for both a two year and a four year degree is also extraordinarily important to institutional funding. Higher education institutions would prefer higher numbers of successful full time students to boost their retention and completion rates. In a study of community college students, Yu (2017) found that full time attendance increased the likelihood of persistenc e and completion. The time it takes to complete a degree is reduced, accrual of credit happens at a faster rate, and external pressures may be lessened. Attewell and Monaghan (2016) found that for full time students, increased credit loads had a positive e ffe ct on student academic momentum. Pell Status Receiving a Pell grant was not found to be significant for either first semester or first year retention for the 2013 cohort. This is in opposition to much of the existing literature which has found Pell awar ds being related to increased student persistence and retention (Adelman, 1999; Bettinger, 2004; Cabrera, Nora, & Castenada, 1992). However, in an analysis of first year community college students, Hawley & Harris (2006) found that receipt of a Pell grant was not a strong motivator for either retention nor attrition. If one examines the rules regarding Pell eligibility, these findings are not so unusual. To continue receiving Pell funding, a student must maintain satisfactory academic progress (as determin ed by the individual institution) which includes credit accrual and maintaining a cumulative GPA in line with the hours attempted. However, a 2.00 GPA must be maintained after the second year of college (U.S.DOE, 2017). At the

PAGE 72

72 College, students wishing to continue receiving Pell must pass 67 percent of their coursework each semester as well as maintain a minimum GPA of 2.00. In the regression analysis of first year GPA, Pell was one of the most highly significant indicators of academic success Nearly 21 percent of the Pell recipients who attended both semesters the first year had a college GPA below a 2.00 as compared to less than 4 percent of non Pell recipients Sixty one percent of Pell recipients had a GPA below 3.00 whereas 39 percent of the non Pell recipients were below a 3.00 GPA. When one looks at total number of students who returned in Fall of 2014, ninety one Pell recipients (44 percent of the incoming Fall 2013 cohort) did not return for their second year of college as compared to fifty four non Pell recipients (34 percent). Pell eligibility cannot account for all the attrition of these students based on their college GPA but lack of funding can certainly be attributed to part of the reason for not returning the second year. Prior research of low income students (to which Pell is applicable) has those in higher socio economic strata (Chen & Carroll, 2005). Gender Student gender was not found to have a signifi cant relationship with student retention Adelman (1999) found that gender played no role in student retention and completion in early analyses of the data prior to investigation of credit hour accrual. This lack of significance remained when Adelman (2006 whether the variables of interest in 1999 were still of interest after several educational reforms had been introduced. Gender, by itself, does not significantly contribute to determining likelihood of retention.

PAGE 73

73 Ethnici ty Ethnicity was not found to have a significant impact on student retention; however, there was a slight significance first year college GPA and minority status with white students being more likely to have a higher GPA That significance, however, disappeared with the addition of successful credit accrual. Adelman (1999) also found that minority status had a slightly negative association with later degree attainment until actual student performance i.e. credit accrual, was taken into account. findings While the literature consistently shows that ethnicity influences retention (DeNicco et al., 201 5; Pascarella, Smart & Ethington, 1986), credit accrual was not an included variable for those researchers. Academic Preparedness via HS GPA and Retention High school GPA has been used as one of the selective measures for entrance into four year institu tions for decades. It has been less scrutinized at the community college since they are open indicative of how successful a student is likely to be in college. Therefore, the significance of HS GPA in relationship with retention for the Fall 2013 cohort was not surprising. As expected, HS GPA was a significant predictor of cohort retention based on both individual high risk class effects as well as first semester outcomes. Moreover, there was a high correlation between HS GPA and first year college GPA. Many researchers (Adelman, 1999; DesJardins et al., 2002; Feldman, 1993; Ishitani, 2006) have found that high school performance was directly related to college success. A recent study of FTIC students entering the University of Alaska

PAGE 74

74 (Hodara & Cox, 2016, p. 10). increase in HS GPAs (moving from a 2.0 to a 3.0, for example) increased the likelihood of passing college level English and math by more than 25 percentage points. The decreased relationship between HS GPA and retention is notable in this study for the students that remained enrol led for back to back semesters. DesJardins et al., (2002) noted that the effect of high school performance decreased with each college performance is controlled for, the strength of pre college academic measures becomes the most significant indicator for continued college retention. For the FTIC Fall 2013 cohort, the college GPA becomes a b etter measure of retention after the first year of enrollment. Academic success via college GPA and Retention The impact of college GPA on retention was not unexpected. The first semester GPA is the first indication of long term success in college and was the most significant indicator of retention over all other indicators, including HS GPA. The average first semester GPA was a 2.88. For students attending two consecutive semesters, the average GPA at the end of the first year was a 2.82. However, the ave rage GPA for students returning the second year was a 3.03 This correlates with Leinbach & Jenkins (2008) study which found that students with a GPA of 3.0 or higher were more likely to be retained. For members of the Fall 2013 cohort with GPAs below a 2 .00, the rate of return was less than five percent. Overall, 45.21 percent of the returning Fall 2013 FTIC

PAGE 75

75 cohort had GPAs below a 3.00, but well over half of those returning earned a GPA between 2.50 and 2.99. Accruing Credit The results of this study clearly demonstrate the correlation between accruing credit hours and student retention. A larger percentage of students with greater than 20 credit hours (64.62%) returned for the second year of college as opposed to only 35.38 percent of the students who earned less than 20 credit hours the first two semesters. The success rate (percentage of credit hours earned versus credit hours completed) with which students achieved their credit accrual is very telling in the student retention rates. Students who ac hieved between 75 percent to 100 percent of their attempted hours accounted for 71.69 percent of the total number of FTIC students returning for a seco nd year. Of those 152 students, 103 were 100 percent success ful in their attempted classes and 92 of the 103 earned more than 20 hours. An examination of success rate in correlation with first year GPA also shows that most students who returned the second year had GPAs above a 2.00. Moreover, a greater number of these had earned 20 or more credit hours. Ther e were a few students with high GPAs and low credit hour accrual that remained for the second year which is expected of part time students. However, those with both low credit hours earned and low GPAs were, for the most part, absent at the beginning of th e second year. This pattern of retention or attrition aligns with the concept of academic momentum. Students who accrue credit at a steady pace will, for the most part, remain in college whereas those who are less successful will drop out.

PAGE 76

76 Implications for Institutional Practice Sixty percent of the Fall 2013 FTIC cohort remained at the College after the first year. While this is laudable, the loss of forty percent of the cohort is not. Some of these non returning students may have transferred to other in stitutions and some may have decided that their career pathway did not involve an AA/AS degree. The students who dropped out because of poor performance, however, are the ones that can best be served by the information gleaned from this study. There have been numerous studies identifying gateway/gatekeeper courses as direct impediments to student success in college (Buchanan 2006; Calcagno et al. 200 7 ; Kuh, 2001; Roksa et al., 2009 ) One of the most difficult subject areas in which to accrue credit is mat hematics, especially in developmental math and introductory algebra. Kuh (2001) reported that 30 40 percent of students enrolled in a gateway mathematics course at a four year residential university either earned a D, an F or withdrew from the course. For the College, MAT 1033 is a very high risk gateway course which inevitably has large percentages of students in the D/F/W category. The developmental education opt out clause for Florida public high school graduates and military personnel poses a significa nt challenge in providing interventions to prevent dropping out from MAT 1033 which may then prevent attrition from the College. Providing a complete picture of the risks associated with taking MAT 1033, especially for those students with low HS GPAs and/ or part time enrollment, could assist them in making better decisions regarding scheduling of classes. For FTIC students who need developmental math but choose to opt out, additional ammunition in the toolbox can only help student advisors. Oftentimes FTIC students are unaware of their deficiencies in classes like MAT 1033 even when placement scores are in front of them. Park et al.,

PAGE 77

77 (2016) posit that FTIC students may not be able to accurately self place themselves in the correct courses which leads to fa ilure in the gateway course. The College could take advantage of the results of this study to better serve a population of FTIC students who have lofty ideas of their abilities but have low probabilities of succeeding. For STEM students, course taking beha vior and their successful accrual of credit is v itally important. Crisp et al., (2009) noted that the additional challenges posed by introductory math and science gatekeeper courses may lead to disillusionment with STEM courses and subsequently deterring s tudents from that pathway. In an analysis of transcript data from the Postsecondary Education Transcript Study (PETS:09) coupled with data from the Beginning Postsecondary Students Longitudinal Study (BPS:04/09), Wang (2016) examined the course taking pat terns of community college students and their subsequent transfers to four year colleges as it related to STEM. One of the outcomes of this analysis was that first semester and first year course taking choices were instrumental in continued enrollment in a STEM pathway. Lack of forward momentum in the math and science courses, or delay in taking gateway math and science courses, led to students leaving the STEM pathway or being unsuccessful in their chosen STEM field of study after transferring to a four ye ar university. The College should monitor these students closely, especially in the first semester, so that needed interventions can be encouraged by advisors and faculty. Part time attendance is a necessity for some students However, starting college as a part time attendee poses a pro blem. Part time attendance has both a negative effect on retention and degree completion (Adelman, 1999, 2006; Byun, Irvin, & Meece, 2012) Attwell et al., (2012) noted that first time students who enroll part time tend t o be

PAGE 78

78 weaker academically and come from a lower SES background. These students tend to consistently lag behind their peers throughout their academic ca reers and are less likely to persist. A study of course completion and time to degree by Kress (2007) found that a majority of part time FTIC students were un successful in nearly 40 percent of their coursework. This inability to successfully complete credits extends the time to degree ev en further and greatly increases the likelihood of dropping out. For t he College, focusing on the incoming part time FTIC students during their first semester could be advantageous for student retention. A study by MacCann, Fogarty, & Roberts (2011) found that time management was extraordinarily important for part time commu nity college students to be successful in college. Since many part time students are juggling jobs, families and coursework, the College could institute strategies to help these students better manage their study time. The College currently does not utili ze HS GPAs as a measure of student ability during the advising process. The outcomes of this research indicate that the College may want to use HS GPA as an additional piece of information to better assist student scheduling. Students with lower HS GPAs m ay need to be tracked more closely during their first semester so that interventions can be undertaken before course failure occurs. Tucker and McKnight (2017) found that students with HS GPA below a 2.3 were more likely to have a low college GPA and be l ess likely to be successful academically. Likewise, Yu (2017) noted that college completion in a timely manner was positively correlated with high school GPA. There are concerns, however, as to the veracity of HS GPAs due to grade inflation. A recently r eleased study cited by Jaschik (2017), in which HS GPAs were

PAGE 79

79 examined between the years of 1998 and 2016, found the average HS GPA to have risen from 3.27 to 3.38. However, these grade inflations were not uniformly disbursed. Wealthier schools, with a grea ter percentage of whites enrolled, were found to have larger increases than lower income, higher minority schools. While the College is in an area of low minority populations, the income of the rural population is most certainly not high. Grade inflation m ay not be an issue within this define d rural area but caution would be a dvisable. If the College were to use student HS GPA s as indicator s of potential ability in gateway classes, advisors would need to balance it wi th other available information. The co rrelation between first semester college GPA and student retention has been well documented and the results of this study align with the literature. Kahn and Nauta (2001) found that first semester GPA was the strongest predictor of student persistence into the sophomore year in their study of 400 freshmen students whose gender and ethnicity make up were very similar to the students at the College. Gershenfeld et al., (2016) noted that a first semester GPA of 2.33 or below reduced the odds of graduating by n early half. Identifying students with low first semester GPAs and discussing strategies to assist them in obtaining better outcomes could benefit both the students and the College. Early intervention may reduce D/F/W rates as well as increase student rete ntion rates. An analysis of data from BPS:04/09 found that r egardless of the type of four year institution (urban, suburban or rural) first time college students with a first year GPA of 3.5 were 47 percent more likely to continue their college educatio n than those with a GPA below 3.5 (Sparks and Nunez, 2014). DesJardins and McCall (2010) noted for

PAGE 80

80 each full increased by nearly 30 percent Achieving a strong college GPA in the first semester is a positive experience for students which encourages them to continue their college career. Moreover, a high GPA is indicative of successfully passing courses and accruing credits which is a hallmark of academic momentum. The College in th is study has recently added a se rvice called Dropout (personal knowledge) which tracks student performance in their classes. Advisors are alerted when students have low class performance (as measured by grade inputs and class attendance), and fa culty can alert advisors as well. However, the success of the program is dependent on timely grade and attendance inputs by faculty and continuous monitoring by advisors. Even with the addition of Dropout Detective higher than desired D/F/W rates are still occurring. By using the information from this study, the College could implement a multi pronged strategy involving pre semester advising, in semester tracking with Dropout Detective and post semester follow up with students who fall below a 2.50 GPA or have a percentage accrual of credit below the 74 percent range. Pre semester advising for FTIC students should include a discussion of HS GPA as a measure of a cademic readiness and how it relates to first semester success. Tracking students during t he semester can allow for timely interventions which may prevent student withdrawals or failures which slow progress and, in cases of low grades, lower the ir college GPA which may impact their standing with financial aid Finally, students who are not main taining a 2.50 GPA are at greater risk, based on the results of both this study and prior research, of dropping out of college. Failure to maintain a 75 percent or greater credit accrual

PAGE 81

81 rate, whether full time or part time, diminishes the likelihood of st udent persistence and ultimately completion of their college degrees. These strategies could increase both student achievement in the classroom and student retention rates for the College. As Witteveen & Attewell (2016) so succin c tly say, For average Ame rican college students, academic momentum may taking pattern in the first couple of semesters. This emphasizes the importance of careful course selection a responsibility of st udents themselves, but also of administr ators, faculty, and counselors. (p p .463 464) Recommendations for Further Research This study provides insight into characteristics of students at a small, rural community college which can directly impact student ret ention. The use of a single cohort at one institution cannot be generalized for all rural institutions much less for large urban community colleges Utilizing data from several different cohorts can potentially paint a fuller picture as to the impact cours e taking behavior has on student success. Additionally, the study did not disaggregate FTIC from first generation in college (FGIC) students. Research into FGIC students has shown these students do exhibit specific needs that may impact their college succe ss over and above those discussed within this study. Credit accrual and the methods by which it can be accrued is an area which could be addressed as well. This study does not differentiate between courses which were delivered online versus face to face, nor does it investigate the impact of dual enrollment credit on student success. and counselors have to target every first time student for at least 20 additive credits by the end of the first calendar year of enrollment p. 109). In his opinion, even six hour s

PAGE 82

82 of dual enrollment credit prior to entering college would alleviate the problems of part time a ttendees lack of timely credit accrual the first year. These areas are ripe for future research since more students are taking advantage of both dual enrollme nt opportunities and online availability of credit bearing courses This study focused on a small subset of data to which the researcher had access. Previous research has shown that social integration plays a significant role in student retention. Additi on of student engagement variables, such as participation in college activities, utilizing tutoring services and family characteristics, would add to the richness of the study and provide insight into both non cognitive and non demographic factors which do have an impact on student retention. These factors, especially within the rural context, would be of great assistance to institutions in deciding how best to address their student retention activities to minimize drop out risk. Limitations This study is m arked by several limitations. The population studied, while mirroring the communities the College serves, has several differences when compared to the national norms for community colleges. The small number of minority students for example, does not allow for the results of this study to be gene ralized to a larger population, especially one whose demographics are markedly different from those of the Furthermore, disaggregation of the data by gender plus ethnicity cou ld provide greater insight into patterns of retention not examined by this study. V ariables such as age and SES status were not used in this study. The average year universities rath er than the higher averages seen in most community colleges. Age has been shown

PAGE 83

83 to play a role in much of the research but was purposefully omitted in this study. Family income (SES) was not readily available to this researcher due to the way records were kept at the College prior to 2016 Remedial, or developmental, coursework was not included in this study. The cohort in this study would have been required to take developmental education courses if their incoming test scores warranted it. However, the l egislative changes in determining who was required to take developmental education courses, instituted in the state of Florida in 2014, significantly reduced the enrollment of students in these classes which has reduced the impact of these courses on reten tion statewide. Inclusion of these courses would not have assisted in determination of drop out risk for the institution utilized in this study. Additionally, non academic factors were not included as they were beyond the scope of the study Measures of st udent engagement which often include w ork/family obligations, available institutional support services, interactions with faculty, and student participation in campus activities are unaccounted in this limited study Summary of Discussion Research on ru ral institutions in general has been underrepresented in the literature with small rural colleges being investigated even less. Rural demographics have been shown to be, in many cases, markedly different than those found in urban areas. Generalizations as to drop out behavior, based on research at large urban institutions, may not be applicable to the rural institutions, especially those that are small. The objecti ve of this study was to explore the relationship of high risk courses to first year drop out risk for incoming first time in college students at a small rural community college. The focus of the model was the impact of high risk courses, and the

PAGE 84

84 way they are taken, as sources of drop out risk. Overall, high risk courses cannot be identified as the source of student attrition. They do, however, contribute to student performance which ultimately determines success in college.

PAGE 85

85 A PPENDIX A ADDITIONAL RESULTS Table A 1 Effect of BSC1005 on Fall to Spring retention Variables Odds Ratio Robust Standard Error z Sig. Female 0.746 0.235 0.93 0.353 White 0.621 0.222 1.33 0.183 Part time 0.266 0.090 3.90 0.000*** Pell recipient 1.100 0.338 0.31 0.756 BSC1005 1.757 1.172 0.85 0.398 HS GPA 2.022 0.559 2.55 0.011* Constant 1.133 1.086 0.13 0.896 Note: N=364. *** p Table A 2 Effect of BSC2010 on Fall to Spring r etention Variables Odds Ratio Robust Standard Error z Sig. Female 0.733 0.230 0.99 0.321 White 0.661 0.235 1.16 0.245 Part time 0.263 0.090 3.90 0.000*** Pell recipient 1.075 0.329 0.23 0.815 BSC2010 0.505 0.339 1.02 0.309 HS GPA 2.081 0.580 2.63 0.008** Constant 1.094 1.045 0.09 0.925 Note: N=364. *** p Table A 3 Effect of MAC1105 on Fall to Spring r etention Variables Odds Ratio Robust Standard Error z Sig. Female 0.739 0.232 0.96 0.335 White 0.642 0.226 1.26 0.208 Part time 0.273 0.094 3.76 0.000*** Pell recipient 1.092 0.337 0.29 0.776 MAC1105 1.021 0.398 0.05 0.958 HS GPA 2.015 0.548 2.58 0.010** Constant 1.165 1.114 0.16 0.873 Note: N=364. *** p

PAGE 86

86 Table A 4 Effect of ENC1101 on Fall to Spring r etention Variables Odds Ratio Robust Standard Error z Sig. Female 0.727 0.228 1.02 0.309 White 0.641 0.226 1.26 0.207 Part time 0.276 0.094 3.79 0.000*** Pell recipient 1.097 0.335 0.30 0.763 ENC1101 1.254 0.396 0.72 0.474 HS GPA 2.034 0.561 2.57 0.010** Constant 1.055 1.023 0.06 0.956 Note: N=364. *** p Table A 5 Effect of MUL2010 on Fall to Spring r etention Variables Odds Ratio Robust Standard Error z Sig. Female 0.744 0.235 0.94 0.349 White 0.639 0.227 1.26 0.207 Part time 0.274 0.093 3.80 0.000*** Pell recipient 1.107 0.340 0.33 0.741 MUL2010 1.213 0.703 0.33 0.739 HS GPA 2.004 0.556 2.50 0.012* Constant 1.161 1.119 0.16 0.877 Note: N=364. *** p Table A 6 Effect of CHM1030 on Fall to Spring r etention Variables Odds Ratio Robust Standard Error z Sig. Female 0.734 0.232 0.98 0.329 White 0.647 0.229 1.23 0.217 Part time 0.269 0.092 3.86 0.000*** Pell recipient 1.080 0.331 0.25 0.802 CHM1030 0.591 0.670 0.46 0.643 HS GPA 2.039 0.563 2.58 0.010** Constant 1.147 1.100 0.14 0.886 Note: N=364. *** p

PAGE 87

87 Table A 7 Effect of CHM1045 on Fall to Spring r etention Variables Odds Ratio Robust Standard Error z Sig. Female 0.711 0.226 1.07 0.284 White 0.653 0.231 1.21 0.228 Part time 0.265 0.091 3.87 0.000*** Pell recipient 1.088 0.335 0.27 0.785 CHM1045 0.585 0.350 0.90 0.370 HS GPA 2.147 0.619 2.65 0.008** Constant 1.019 0.992 0.02 0.985 Note: N=364. *** p Table A 8 Effect of ESC1000 on Fall to Spring r etention Variables Odds Ratio Robust Standard Error z Sig. Female 0.739 0.231 0.97 0.334 White 0.642 0.226 1.26 0.208 Part time 0.272 0.092 3.84 0.000*** Pell recipient 1.091 0.335 0.28 0.777 ESC1000 1.033 0.633 0.05 0.957 HS GPA 2.019 0.559 2.54 0.011* Constant 1.162 1.115 0.16 0.875 Note: N=364. *** p Table A 9 Effect of BSC2093 on Fall to Spring r etention Variables Odds Ratio Robust Standard Error z Sig. Female 0.748 0.236 0.92 0.357 White 0.642 0.226 1.26 0.208 Part time 0.270 0.091 3.87 0.000*** Pell recipient 1.089 0.333 0.28 0.781 BSC2093 0.619 0.452 0.66 0.511 HS GPA 2.060 0.576 2.58 0.010** Constant 1.106 1.067 0.10 0.917 Note: N=364. *** p

PAGE 88

88 Table A 1 0 Retention vs p ercentage of successful c redit a ccrual Total Credit Hours Earned <20 h ou rs Percentage Successful Credit Accrual Number Students Retained Number Students Not Retained Number Students Retained Number Students Not Retained 100% 92 19 10 3 75 99% 41 7 8 5 50 74% 4 2 39 13 25 49% 0 0 14 20 0 24% 0 0 3 19 Note: N=364.

PAGE 89

89 APPENDIX B INSTITUTIONAL IRB APPROVALS

PAGE 90

90

PAGE 91

91 LIST OF REFERENCES ACT. (2015). National collegiate retention and persistence to degree rates Retrieved from https://www.ruffalonl.com/documents/shared/Papers_and_Research/ACT_Data/ ACT_2015.pdf ACT, Inc. (2016). The cond ition of college & career readiness 2016 Retrieved from http://www.act.org/content/dam/act/unsecured/documents/CCCR_National_2016. pdf Adelman, C. (1999). Answers in the tool box. Academic intensity, attendance patterns, (PLLI 1999 8021). Washington, DC: Government Printing Office. Adelman, C. (2006). The toolbox revisited: Paths to degree completion from high school through colle ge Retrieved from U.S. Department of Education: http:// www.ed.gov/pubs/edpubs.html Allison, P. (2012) When can you safely ignore multicollinearity? Retrieved from https://statisticalhorizons.com/multicollinearity Alon, S. (2011). Who benefits most from financial aid? The heterogeneous effect of need Social Science Quarterly 92 (3), 807 829. http://dx.doi.org/10.1111/j.1540 6237.2011.00793.x American Association of Community Colleges. (2016). Fast Facts Retrieved from http://www.aacc.nche.edu/AboutCC/Documents/AACCFactSheetsR2.pdf Astin, A. W., & Oseguera, L. (2012). Pre college and institutional influences on degree attainment. In A. Seidman (Ed.), College student retention: Formula for student success (2nd ed. (pp. 119 145). [VitalSource Bookshelf version]. Attewell, P., Heil, S., & Reisel, L. (2011). Competing explanations of undergraduate noncompletion. American Educational Research Journal 48 (3), 536 559. http://dx.doi.org/10.3102/0002831210392018 Attewell, P., Heil, S., & Reisel, L. (2012, March). What is academic momentum? And does it matter? Educational Evaluation and Policy Analysis 34 (1), 27 44. Retrieved from http://www.jstor.org/stable/41413073 Attewell, P., & Monaghan, D. (2016). How many credits should an undergraduate take? Research in Higher Education 57 682 713. http://dx.doi.org/10.1007/s11162 015 9401 z Bailey, T. (2009). Challenge and opportunity: Rethinking the role and function of developmental education in community college. New Directions For Community Colleges 145 11 30. http://dx.doi.org/10.1002/cc.352

PAGE 92

92 Bailey, T., Calcagno, J. C., Jenkins, D., Kienzl, G., & Leinbach, T. (2005). The effects of institutional factors on the success of community college students Retrieved from Community College Research Center website: http://ccrc.tc.columbia.edu/media/k2/attachments/effects institutional factors success.pdf Bailey, T., Calcagno, J. C., Jen kins, D., Leinbach, T., & Kienzl G. (2006). Is student right to know all you should know? An analysis of community college graduation rates. Research in Higher Education 47 (5), 491 519. http://dx.doi.org/ 10.1007/s11162 005 9005 0 Balfanz, R., DePaoli, J.L., Ingram, E.S., Bridgeland, J.M., & Fox, J.H. (2016). Closing the college gap: A roadmap to postsecondary readiness and attainment. Retrieved from Civic Enterprises website: http://www.civicenterprises.net/MediaLibrary/Docs/CCR.pdf Barnett, E. A. (2011). Validation experiences and persistence among community college students. The Review of Higher Education 34 (2), 193 230. http://dx.doi.org/10.1353/rhe.2010.0019 Bean, J. P. (1980). Dropouts and turnover: the synthesis and test of a causal model of student attrition. Research In Higher Education 12 (2), 155 187. Bean, J. P ., & Metzner, B. S. (1985). A conceptual model of nontraditional undergraduate student attrition. Review of Education Research 55 (4), 485 540. Retrieved from http://www.jstor.org/stable/1170245 Bettinger E. (2004). How financial aid affects persistence [Working paper 10242]. Retrieved from National Bureau of Economic Research: http://www.nber.org/papers/w10242.pdf Blanchard, R. A., Casados, F., & Shes ki, H. (2009). All things to all people: Challenges and innovations in a rural community college. The Journal of Continuing Higher Education 57 22 28 h ttp://dx.doi.org/10.1080/07377360902804077 Bowen, W.G., Chingos, M.M., & McPherson, M.S. (2009). Crossing the finish line: University Press. Braxton, J. M. (2014). Rethinking college student retention [MyiLibrary version]. Retrieved from http://lib.myilibrary.com/Open.aspx?id=540314 Brooks Leonard, C. (1991). Demographic and academic factors associated with first to second term retention in a two year college. Community/Junio r College 15 57 69.

PAGE 93

93 Buchanan, D. G. (2006). An exploration of the gateway math and science course relationships in the Los Angeles community college district. (Doctoral Dissertation). UMI Number: 3236483 Byun, S., Meece, J. L., & Irvin, M. J. (2012b). R ural nonrural disparities in postsecondary educational attainment revisited. American Educational Research Journal, 49(3), 412 437. https://doi.org/10.1353/rhe.2012.0023 Cabrera, A. F., Nora, A., & Castaneda, M. B. (1992). The role of finances in the persistence process: A structural model. Research in Higher Education, 33(5), 571 593. Calcagno, J. C., Crosta, P., Bailey, T., & Jenkins, D. (2007). Stepping stones to a degree: The impact of enrollment pathways and milestones on community college students Research in Higher Education, 48(7), 774 801. http://dx.doi.org/10.1007/s11162 007 9053 8 Carnegie Foundation. (2010). Carnegie classificati on Basic classification methodology Retrieved from http://classifications.carnegiefoundation.org/methodology/basic.php Carnegie Foundation for the Advancement of Teachin g. (2006). The Carnegie classification of institutions of higher education [2005 edition]. Stanford, CA: Author. Chen, X., & Carroll, C. (2005). First generation student in postsecondary education: A look at their college transcripts. (NCES No. 2005 171). U.S. Department of Education, National Center for Education Statistics. Retrieved from NCES http://nces.ed.gov/ pubs2005/2005171.pdf Chen, R., & St. John, E. P. (2011). State financial policies and college student persistence: A national study. The Journal of Higher Education 82 (5), 629 660. Retrieved from http://www.jstor.org/stable/29789545 Cohen, B. (2013). Explaining psychological statistics (4th ed.). [VitalSource Bookshelf version]. Common Placement Testing for Public Postsecondary Education, 1008 Fla. Stat. § 30 (2013). Complete College America. (2011). Time is the enemy Retrieved from http: //www. completecollege.org Consolidated Appropriations Act, 2012, H.R. Res. HR 3671, 112th Cong., 1 Congress.gov 1 (2012) (enacted).

PAGE 94

94 Conway, K. M. (2009). Exploring persistence of immigrant and native students in an urban community college. The Review of Higher Education 32 (3), 321 352. http://dx.doi.org/10.1353/rhe0.0059 Crisp, G., & Mina, L. (2012). The community college: retention trends and issues. In A. Seidman, College student retention (2nd ed. (p p. 147 165). [VitalSource Bookshelf version]. Crisp, G., Nora, A., & Taggart, A. (2009). Student characteristics, pre college, college, and environmental factors as predictors of majoring in and earning a STEM degree: An analysis of students attending a H ispanic serving institution. American Educational Research Journal 46 (9), 924 942. Retrieved from http://www.jstor.org/stable/40284742 Crockett, K., Heffron, M., & Schneider, M. (2011) Targeting financi al aid for improved retention outcomes: The potential impact of redistributing state gift aid on regional universities. American Institutes for Research. Retrieved from: http://www.air.org/sites/default/files/downloads/report/LA_PELL_STUDY_report_ 1011_0.pdf Darling, R. A. (2015). Creating an institutional academic advising culture that supports commuter student success. New Directions for Student Services 150 87 96. http://dx.doi.org/0.1002/ss DeNicco, J., Harrington, P., & Fogg, N. (2015). Factors of one year college retention in a pub lic state college system. Research in Higher Education Journal 27 (), 1 13. Retrieved from http://files.eric.ed.gov/fulltext/EJ1056244.pdf DesJardins, S. L ., McCall, B. P., Ahlburg, D. A., & M oye, M. J. (2002, February). Adding Research in Higher Education 43 (1), 83 114. http://dx.doi.org/10.1023/A:1013022201296 DesJardins, S. L., & McCall, B. P. (2010, Summer). Simulating the effects of financial aid packages on college student stopout, reenrollment spells, and graduation chances. The Review of Higher Education 33 (4), 513 541. http://dx.doi.org/10.13 53/rhe.0.0169 Doyle, W. R. (2009). Impact of increased academic intensity on transfer rates: An application of matching estimators to student unit record data. Research in Higher Education 50 52 72. http://dx.doi.org/10.1007/s11162 008 9107 6 Dynarski, S. M. (1999). Does aid matter? Measuring the effect of student aid on college attendance and completion [Working paper 7422]. Retrieved from National Bureau of Economic Research: http://www.nber.org/papers/w7422.pdf

PAGE 95

95 Eagan Jr., M. K., & Jaeger, A. J. (2008, Fall). Closing the gate: Part Time faculty instr uction in gatekeeper courses and first year persistence. New Directions for Teaching & Learning 2008 (115), 39 53. http://dx.doi.org/10.1002/tl.324 Fairchild, E. E. (2003). Multiple roles of adult learners. New Directions for Student Services 102 11 16. Feldman, M. J. (1993). Factors associated with one year retention in a community college. Research in Higher Education 34 (4), 503 512. http://dx.doi.org/ 10.1007/BF00991857 Florida College System. (2016). FCS Facts Retrieved from http://www.fldoe.org/schools/higher ed/fl college system/facts at a glance.stml Flo rida College System Performance Based Incentive, 1001 Fla. Stat. § 66 (2016). Florida College System Performance Based Incentive, XLVIII 2016 Florida Statutes § 66 (2016). Florida Statute 1008.30, SB 1720 FL Rules § 2013 40 (2013). Fluharty, C., & Scaggs, B. (2007). The rural differential: Bridging the resources gap. New Directions For Community Colleges 137 19 26. http://dx.doi.org/10.1002/cc.266 Forman, S. W. (2009). Characteristics of successful communit y college students (Doctoral dissertation). Retrieved from http://search.proquest.com/docview/305093241?pq origsite=summon Freeman, S. F. (2008). The relationship between academic momentum and undergraduate degree completion at one historically black insti tution (Doctoral dissertation). Available from ProQuest Dissertations and Theses database. (3340864) French, B. F., Immekus, J. C., & Yen, H. (2013). Logistic regression. In T. Teo (Ed.), Handbook of quantitative methods for educational research (pp. 145 1 66). [Adobe Digital Editions version]. http://dx.doi.org/10.1007/978 94 6209 404 8 Gainen, J. (1995). Barriers to success in quantitative gatekeeper courses. New Directions for Teaching and Learni ng 61 (), 5 14. http://dx.doi.org/10.1002/tl.37219956104 Gershenfeld, S., Hood, D. W., & Zhan, M. (2016). The role of first semester GPA in predicting graduation rates of underrepresented students. J ournal of College Student Retention: Research, Theory & Practice 17 (4), 469 488. http://dx.doi.org/10.1177/1521025115579251

PAGE 96

96 Goldrick Rab, S. (2007). Promoting academic momentum at community colleges: Challenges and opportunities (CCRC Working Paper No. 5). New York, NY: Columbia University, Community College Research Center. Habley, W. R., Bloom, J. L., & Robbins, S. (2012). Increasing Persistence: Research based strategies for college stude nt success San Francisco, CA: Jossey Bass. Habley, W., Valiga, M., McClanahan, R., & Burkum, K. (2010). What works in student retention? Fourth national survey Retrieved from ACT website: http:// www. act.org Hawley, T. H ., & Harris, T.A. (2005). Student characteristics related to persistence for first year community college students. Journal of College Student Retention, 7 (1 2), 117 142. Hardy, D. E., & Katsinas, S. G. (2007, Spring). Classifying community colleges: How rural community colleges fit. New Directions for Community Colleges 137 5 17. http://dx.doi.org/10.1002/cc.265 Health and Human Services. (2017). Annual update of the HHS poverty guidelines Retrieved from : https://www.federalregister.gov/documents/2017/01/31/2017 02076/annual update of the hhs poverty guidelines Hicks, C., & Jones, S. J. (2011). At Issue: survival tactics for small, rural serving community colleges. The Community College Enterprise 17 (2), 28 45. Retrieved from http://search.proquest.com/do cview/912388505?pq origsite=summon Hodara, M., & Cox, M. (2016). Developmental education and college readiness at the University of Alaska (REL 2016 123). Washington, DC: U.S. Department of Education, Institute of Education Sciences, National Center for Ed ucation Evaluation and Regional Assistance, Regional Educational Laboratory Northwest. Retrieved from http://ies.ed.gov/ncee/edlabs Hoffman, J. L., & Lowitizi, K. E. (2005). Predicting college success with high school grades and test scores: Limitations for minority students. Review of Higher Education 28 (4), 455 474. Hu, S., & Ma, Y. (2010). Mentoring and student persistence in col lege: A study of the Washington State Achievers program. Innovative Higher Education 35 329 341. http://dx.doi.org/10.1007/s10755 010 9147 7 Hu, S., McCormick, A. C., & Gonyea, R. M. (2012). Exa mining the relationship between student learning and persistence. Innovative Higher Education 37 387 395. http://dx.doi.org/10.1007/s10755 011 9209 5

PAGE 97

97 Institute for Digital Research and Educatio n (IDRE). (2107). Regression with STATA chapter 2 regression diagnostics Retrieved from: https://stats.idre.ucla. edu/stata/webbooks/reg/chapter2/stata webbooksregressionwith statachapter 2 regression diagnostics/ Ishitani, T. T. (2006). Studying attrition and degree completion behavior among first generation college students in the United States. The Journal of Highe r Education 77 (5), 861 885. http://dx.doi.org/10.1353/jhe.2006.0042 Jackson, J., & Kurlaender, M. (2014). College readiness and college completion at broad access four year institutions. American Beh avioral Scientist 58 (8), 947 971. http://dx.doi.org/ 10.1177/0002764213515229 Jamelske, E. (2009). Measuring the impact of a university first year experience program on student GPA and retentio n. Higher Education 57 373 391. http://dx.doi.org/10.1007/s10734 008 9161 1 Jaschik, S. (2017). High school grades: Higher and higher. Inside Higher Ed July 17, 2017. Retrieved from: https://www.insidehighered.com/admissions/article/2017/07/17/study finds notable increase grades high schools nationally Jenkins, D., Jaggars, S. S., & Roksa, J. (2009). Promoting gatekeeper course success among community college students needing remediation: Findings and recommendations from a Virginia study Retrieved from Community College Research Center website : http://ccrc.tc.columbia.edu/publications/gatekeeper course success virginia.html Juszkiewicz, J. (2015). Trends in communit y college enrollment and completion data, 2015 Retrieved from American Association of Community Colleges website: http://www.aacc.nche.edu/Publications/Reports/D ocuments/CCEnrollment_2015. pdf Kahn, J. H., & Nauta, M. M. (2001). Social cognitive predictor of first year college persistence: The importance of proximal assessment. Research in Higher Education 42 (6), 633 652. http://dx.doi.org/10.1023/A:1012225510213 Katsinas, S. G., Davis, J. E., Friedel, J. N., Koh, J. P., & Grant, P. D. (2013). The impact of new Pell grant restrictions on community colleges: A three state study of Alabama, Arkansas, and Miss issippi Retrieved from The University of Alabama Education Policy Center: uaedpolicy.ua.edu/uploads/2/1/3/2/21326282/new_three_state_study_.pdf Katsinas, S. G., & Moeck, P. (2002). The digital divide and rural community colleges: Problems and prospects. C ommunity College Journal of Research and Practice 26 (3), 207 224. http://dx.doi.org/10.1080/106689202317245419

PAGE 98

98 Keels, M. (2013). Getting them enrolled is only half the battle: College success as a function of race or ethnicity, gender, and class. American Journal of Orthopsychiatry 83 (2 3), 310 322. http://dx.doi.org/10.1111/ajop.12033 Khuong, H. (2014). Evaluation of a conceptual model of stu dent retention at a public urban commuter university (Doctoral dissertation). Retrieved from http://ecommons.luc.edu/luc_diss/1092 Kress, A. M. (2007). Factors influencing course completion among first time in college associate of arts students in the community college (Doctoral dissertation. Retrieved from http://etd.fcla.edu/UF/UFE0019260/kress_a.pdf Kuh, G. D. (2001). Organizational culture and student persistence: Prospects and puzzles. Journal of C ollege Student Retention 3 (1), 23 39. http://dx.doi.org/10.2190/U1RN C0UU WXRV 0E3M Kuh, G. D. (2009). What student affairs professionals need to know about student engagement. Journal of College Student Development 50 (9), 683 706. http://dx.doi.org/10.1353/csd.0.0099 Kuh, G. D., Cruce, T. M., Shoup, R., Kinzie, J., & Gonyea, R. M. (2008). Unmasking the effects of student engagement on f irst year college grades and persistence. The Journal of Higher Education 79 (5), 540 563. Retrieved from http://www.jstor.org/stable/25144692 Kuh, G. D., Kinzie, J., Buckley, J. A., Bridges, B. K., & Ha yek, J. C. (2006). What matters to student success: A review of the literature Retrieved from NPEC website: http://nces.ed.gov/npec/pdf/kuh_team_report.pdf Laskey, M. L., & Hetzel, C. J. (20 11). Investigating factors related to retention of at risk college students. The Learning Assistance Review 16 (1), 31 43. Lotkowski, V. A., Robbins, S. B., & Noeth, R. J. (2004). The role of academic and non academic factors in improving college retention Retrieved from ACT website: http://files.eric.ed.gov/fulltext/ED485476.pdf Lumina Foundation. (2016). Goal 2025 [Strategic Plan]. Retrieved from https://www.luminafoundation.org/goal_2025 MacCann, C., Fogarty, G.J., & Roberts, R.D. (2012). Strategies for success in education: Time management is more important for part time than full time community college students. Learning and Indivi dual Differences, 22 (5) 618 623. https://doi.org/10.1016/j.lindif.2011.09.015 McIntosh, M. F., & Rouse, C. E. (2009). The other college: Retention and completion rates among two year college students Retrieved from Center for American Progress website: https://cdn.americanprogress.org/wp content/uploads/issues/2009/02/pdf/two_year_col leges.pdf

PAGE 99

99 Milem, J. F., & Berger, J. B. (1997). A modified model of college student persistence: theory of student departure. Journal of College Student Development 38 (4), 387 400. Retrieved from http://search.proquest.com/docview/1416093041?accountid=10920 Miller, M. T., & Tuttle, C. C. (2007). Building communities: How rural community colleges develop their communities and the people who live in them. Community College Journal of Research and Practice 31 (2), 117 127. http://dx.doi.org/10.1080/10668920500441689 Moore, C., & Shulock, N. (2009). Student progress toward degree completion: Lessons from the research literature Retrieved from The Institute for Higher Education Leadership & Policy websit e: http://www.csus.edu/ihelp Morrison, L., & Silverman, L. (2012). Retention theories, models, and concepts. In A. Seidman, College student retention: Formula for student success (2nd ed. (pp. 61 80). [VitalSource Bookshelf version]. Musoba, G. D., & Krichevskiy, D. (2014). Early coursework and college experience predictors of persistence at a Hispanic serving institution. Journal of Hispanic Higher Education 13 (1), 48 62. http://dx.doi.org/10.1177/1538192713513463 Nation Center for Children in Poverty. (2014). United States demographics of low income children Retrieved from http://www.nccp.org /profiles/US_profile_6.html National Center for Education Statistics. (2016 a ). Characteristics of postsecondary students Retrieved from https://nces.ed.gov/programs/coe/indicator_csb.asp N ational Center for Education Statistics. (2016 b ). Fast facts: Degrees conferred by race and sex Retrieved from https://nces.ed.gov/fastfacts/display.asp?id=72 National Conference of State Leg islatures website. (2015). http://www.ncsl.org/research/education/performance funding.aspx National Student Clearinghouse Research Center. (2014). Snapshot report: persistence retention Retrieved from http://research.studentclearinghouse.org Nora, A., & Crisp, G. (2012). Student persistence and degree attainment beyond the first year in college: Existing knowledge and di rections for future research. In A. Seidman (Ed.), College Student Retention: A Formula for Student Success (2nd ed. (pp. 229 249). [VitalSource Bookshelf version]. Offenstein, J., Moore, C., & Shulock, N. (2010). Advancing by degrees: A framework for i ncreasing college completion Retrieved from The Education Trust website: http://www.edtrust.org

PAGE 100

100 Park, T., Woods, C. S., Richard, K., Tandberg, D., Hu, S., & Jones, T. B. (2016). When developmental education is optional, what will students do? A preliminary analysis of survey data on student course enrollment decisions in an environment of increased choice. Innovative Higher Education 41 221 236. http://dx.doi.org/10.1007/s10755 015 9343 6 Pascarella, E. T., Smart, J.C., & Ethington, C.A. (1986). Long term persistence of two year college st udents. Research in Higher Education, 24 (1), 47 71. http://dx.doi.org/10.1007/BF00973742 Pascarella, E. T., & Terenzini, P. T. (1980). Predicting freshman persistence and voluntary dropout decisions from a theoretical model. The Journal of Higher Education 51 (1), 60 75. Retrieved from http://www.jstor.org/stable/1981125 Pennington, K., Williams, M. R., & Karvonen, M. (2006). Challenges facing rural comm unity colleges: Issues and problems today and over the past 30 years. Community College Journal of Research and Practice 30 (8), 641 655. http://dx.doi.org/10.1080/10668920600746086 Pike, G. R., K uh, G. D., & McCormick, A. C. (2011). An investigation of the contingent relationships between learning community participation and student engagement. Research in Higher Education 52 300 322 http://dx.doi.org/10.1007/s11162 010 9192 1 Ratcliffe, M., Burd, C., Holder, K., & Fields, A. (2016). Defining rural at the U.S. census bureau: American community survey and geography brief Retrieved from U.S. Cen sus Bureau website: http://www2.census.gov/geo/pdfs/reference/ua/Defining_Rural.pdf Reason, R. D. (2003). Student variables that predict retention: Recent research and new dev elopment. NASPA Journal 40 (4), 172 191. Robbins, S., Lauver, K., Le, H., Davis, D., Langley, R., & Carlstrom, A. (2004). Do psychosocial and study skill factors predict college outcomes? A meta analysis. Psychological Bulletin 130 261 288. http://dx.doi.org/10.1037/0033 2909.130.2.261 Rodriguez, D. B. (2013). The influence of student demographics and internal characteristics on GPA, persistence, and academic success of first time college stude nts (Doctoral dissertation). Retrieved from http://search.proquest.com/docview/1492361901 Roksa, J., Jenkins, D., Jaggars, S. S., Zeidenberg, M., & Cho, S. (2009). Strategies for promoting gatekeeper course success among students needing remediation: Research report for the Virginia community college system Retrieved from Community College Research Center website: http://ccrc.tc.columbia.edu Rura l Community College Alliance Website. (n.d.). http://www.ruralccalliance.org/

PAGE 101

101 Schmid, C., & Abell, P. (2003). Demographic risk factors, study patterns, and campus involvement as related to student success am ong Guilford Technical Community College students. Community College Review 31 (1), 1 16. Retrieved from http://dx.doi.org/10.1177/009155210303100101 Scholarships.com. (2017). The Federal Pell Gr ant. Retrieved from https://www.scholarships.com/financial aid/grants/federal grants/ Schuetz, P. (2005). UCLA community college review: Campus environment: A missing link in studies of community college attrition. Community College Review 32 (4), 60 80. Schuetz, P. (2008). A theory driven model of community college student engagement. Community College Journal of Research and Practice 32 (4 6), 305 324. http://dx.doi.org/10.1080/10668920701884349 Smith, W., Droddy, J., & Guarino, A. J. (2011, May 28). Do schools make a difference? A study of high school effects and first year college success. Current Issues in Education 14 (2), 1 7. Spady, W. G. (1970). Dropouts from higher education: An interdisciplinary review and synthesis. Interchange 1 (1), 64 85. http://dx.doi.org/10.1007/BF02214313 Sparks, P. J., & Nune z, A. (2014). The role of postsecondary institutional urbanicity in college persistence. Journal of Research in Rural Education 29 (6), 1 19. Retrieved from http://jrre.vmhost.p su.edu/wp content/uploads/2014/09/29 6.pdf StataCorp. (2015). Stata: Release 14. In Stewart, S., Lim, D. H., & Kim, J. (2015). Factors influencing college persistence for first time students. Journal of Developmental Educat ion 38 (3), 12 20. Stoltzfus, J. C. (2011, October 13). Logistic regression: A brief primer. Academic Emergency Medicine 18 (11), 1099 1104. http://dx.doi.org/10.1111/j.1553 2712.2011.01185.x Thornton, Z. M., & Friedel, J. N. (2016). Performance based funding: State policy influences on small rural community colleges. Community College Journal of Research and Practice 40 (3), 188 203. http://dx.doi.org/10.1080/10668926.2015.1112321 Tinto, V. (1975, Winter). Dropout from higher education: A theoretical synthesis of recent research. Review of Educational Research 45 (1), 89 125. Retri eved from http://www.jstor.org/stable/1170024 Tinto, V. (1993). Leaving college: Rethinking the causes and cures of student attrition (2nd ed.). Chicago; London: University of Chicago Press.

PAGE 102

102 Tinto, V., & Pusser, B. (2006). Moving from theory to action: Building a model of institutional action for student success Retrieved from National Postsecondary Education Cooperative website: https://nces.ed.gov/npec/ Tucker L., & McKnight, O. (2017). Assessing the validity of college success indicators for the at risk student: Toward developing a best practice model. Journal of College Student Retention: Research, Theory & Practice 1 19. http://dx.doi.org/10.1177/1521025117696822 U.S. Census Bureau. (2017). Quick Facts. Retrieved from https://www.census.gov/quickfacts U.S. Department of Agriculture. (2016). Rural cla ssifications Retrieved from https://www.ers.usda.gov/topics/rural economy population/rural classifications/ U.S. Department of Education. (2011). Persistence a nd attainment among Pell grant recipients: Results from the beginning postsecondary students longitudinal study of 2004/09 (NCES 2011 275). Washington, DC: Government Printing Office. U.S. Department of Education. (2017) Federal Student Aid website. Retrie ved from https://studentaid.ed.gov/sa/eligibility/staying eligible Voorhees, R. A. (1987). Toward building models of community college persistence: A logit analysis. Research in Hig her Education 26 (2), 115 129. http://dx.doi.org/10.1007/BF00992024 Walpole, M. (2003). Socioeconomic status and college: How SES affects college experiences and outcomes. The Review of Higher Education 27 (1), 45 73. Retrieved from http://muse.jhu.edu/article/46608 Wang, X. (2016). Course taking patterns of community college students beginning in STEM: Using data mining techniques to reveal viable STEM transfer pathways. Research in Higher Education 57 544 569 http://dx.doi.org/10.1007/s11162 015 9397 4 White House Office of the Press Secretary (Producer). (2009, July 14). Remarks by the President on the American graduation initiative Retrieved from https://www.whitehouse.gov/the press office/remarks president american graduation initiative warren mi Williams, R. (2015). Heteroskedasticity. Retrei ved from https://www3.nd.edu/~rwilliam/stats2/l25.pdf Windham, M. H., Rehfuss, M. C., Williams, C. R., Pugh, J. V., & Tincher Ladner, L. (2014). Retention of first year community college students Community College Journal of Research and Practice 38 466 477. http://dx.doi.org/10.1080/10668926.2012.743867

PAGE 103

103 Witteveen, D. & Attewell, P. (2017) The college completion puzzle: A hidden M arkov model approach. Research in Higher Education, 58 449 467. http://dx.doi.org/ 10.1007/s11162 016 9430 2 o persistence. The Journal of Higher Education 83 (6), 795 823. Retrieved from https://muse.jhu.edu/article/488571 Yu, H. (2017). Factors associated with student academic achievement at community colleges Journal of College Student Retention: Research, Theory & Practice 19 (2), 224 239. http://dx.doi.org/10.1177/1521025115612484 Zajacova, A., Lynch, S. M., & Espenshade, T. J. (2005). Self efficacy stress, and academic success in college. Research in Higher Education 46 (6), 677 706. http://dx.doi.org/10.1007/s11162 004 4139 z

PAGE 104

104 BIOGRAPHICAL SKETCH Terolyn Lay was born in Huntsville, Alabama. She holds a Bachelor of Science degree in m icrobiology from the University of Louisiana, a Master of Science degree in b and a Doctor of Education degree in higher e ducation a dministration from the University of Florida She has held positions in both industry and academia. She is currently an Assistant Professor of b iology at Chipola College in Marianna, Florida where she recently served as the chair of the Quality Enhancemen t Plan team for SACSCOC reaffirmation While a doctoral student, Ms. Lay was awarded the James L. Wattenbarger scholarship from the University of Florida College of Education. She was also a recipient of a SACSCOC Travel Gr ant from SACSCOC to attend the 2 016 SACSCOC annual meeting.